Energy Consumption and Potentials for Energy ...

Energy Consumption and Potentials for Energy Efficiency Implementation: Analyzing Low Income Low Service Areas of Kosovo 2015

Energy consumption and Potentials for Energy Efficiency Implementation: Analyzing Low Income, Low Service Areas of Kosovo – Final Report

December 2015

Authors: Alicia English, Ph.D. Anila Qehaja Flamur Breznica Arif Hoti Michael Waschak, Ph.D. Jim Myers, Ph.D.

12/3/15

Trends and Potentials for LPLS Households

P.1

Acknowledgements We would like to thank the students from the KEDS Academy/ University of Prishtina for their help in collecting the data for the project. The Rochester Institute of Technology in Kosovo (RIT) students that contributed directly to the work in various aspects of this project include Dina Vllasaliu for pre-testing the survey instrument, Blend Hyseni for work on the literature review, Arif Hoti, Anila Qehaja and Flamur Breznica for help writing the report and working on the statistics. Additional appreciation goes to Mr Petrit Papaj and Ms. Iliriana Hajdari from KESCO and Mr. Alper Erbas and Mr. George Karagutoff from KEDS, for providing data and support in regard to the customers and areas of low incomes in Kosovo and organizing this learning opportunity for students to look at issues that affect the country deeply. Funding Statement: This project was funded by the Kosovo Electricity Distribution Supply Company (KEDS), the project monies went to funding research for students at the RIT Center for Energy and Natural Resources. The Principal Investigator on the project, Dr. Alicia English was funded through the Rochester Institute of Technology (RIT) for the work, as there would be less conflict of interest regarding this sensitive topic. Data collection and protection of sensitive data: All researchers who worked on this study were given human subjects research training and the survey was submitted to RIT’s IRB Human Subjects Research Board. In order to protect participant confidentiality, data that could be used to identify participants was redacted from the dataset and ID numbers were assigned.

The Center for Energy and Natural Resources

12/3/15


P.2

SUMMARY OF FINDINGS This report describes the energy situation in households in low payment and low-service areas of Kosovo. The survey explored many different aspects of household interactions with the energy they consume. Participants were asked about their energy expenditures, heating sources, electric appliances, personal behaviors and opinions, many subjects were covered in depth. Below are some of the key findings in each of the areas by section. Demographics -

As typically with rural Kosovo homes, the average number of people from the survey, living within the home is 7 people, though larger families were not uncommon, with 79 households having more than 10 people. This is critical as it impacts not just consumption-based appliances which affect the bill more than larger fixed rate appliances (e.g. refrigerators), but also surface area, income and economic support of members outside the home.

-

Lack of education, and the economic opportunities greater education can provide, is an issue for our sample population; with a majority of adults in the households completing only a high school education. This limits employment opportunities as well as their understanding of both consumption activities and the complexities of the electricity bills.

-

The sources of income for households varied, though most frequently at least one member was employed through full-time jobs, pensions, agriculture/ livestock or receiving remittances. Indicating that when paying monthly bills the frequency of income may be an issue. Those that had a monthly income were reported between €100 and €300 monthly on average and remittances were likely to only be requested as needed.

Energy Expenditure -

One of the statistically relevant factors contributing to higher bills was using their meter for economic activities in addition to household demand. The 29 households indicated that they use their energy meters for economic activities in the households, are subsidizing these activities by €13 to €30 per month.

-

Income is not a key driver in electricity consumption, as households tend to substitute to other energy sources at higher income levels.

12/3/15


P.3

-

The average monthly expenditure on electricity was €43.30 and for wood fuels, €46.74. As can be seen in the section on the expenditure, the mix of fuels varies throughout the sample. The average share of the month’s electrical energy consumption is 34%.

-

The average household consumes 1,191 kWh of non-electric energy sources per month in the winter. The upper 75% quartile consumed 1,447 kWh and the lower 25% of the sample consumed 723 kWh monthly.

-

Consumers are price sensitive to which energy source is used. An increase in electricity price, from one tariff block to the next, results in a 58% decrease in electricity as a share of energy demanded, indicating that households are more flexible in switching energy sources. o The share of electricity that could have been consumed at the higher price is replaced by a 7% increase in wood fuels, and 18% increase in lignite quantity demanded.

-

One significant issue is that the usage-costs for alternatives to electricity are fixed prior to the heating season. The stock of wood fuels and lignite is pre-purchased and prices don’t follow expected demand responsiveness (e.g. higher prices closer to the season, location, etc.) and quantities purchased are not adjusted in the same manner as electricity.

-

Thirty-one percent of the households in our sample would qualify for energy social assistance at 1.72 per person per day income levels, though are currently not on it.

-

Additionally, subsidizing the first 400 kWh may not be the most efficient policy. When calculating out the kWh difference on average, to raise the households to same level as those that would not qualify is a difference of 85 to 125 kWh monthly.

-

When correlating the factors with a house having debt several interesting connections were found. Those that were positively correlated with debts were o The total number of people in the house, o the number of people under the age of 18, o if their main source of income was social assistance, o if their walls or roofs were insulated, o if there were leaks around their windows, o the size of the electricity bill for October and March, o if they communicated with KEDS employees when meters were being read, o if they didn’t pay their bills at the KESCO windows and o those that have problems paying their water and electric bills.

12/3/15


P.4

More explanation is given in the document but these are areas in which more coordinated efforts may be useful in monitoring the progression of households into debt. Heating -

Most homes only heat 41% of their space. Homes are often in the 15 age range, and 367 households have neither roof nor wall insulation. In the preliminary findings, isolation does not appear to be correlated with space heated, as households without isolation have an average of 40% of space heated and the houses with either wall, roof, or both, are only slightly higher with 43%, 44%, 44% heated on average, respectively. It may be likely that the rebound effect on consumption will be small, until the other constraining factors, like income, improve.

-

Several (n:269) households reported that they have air leakage around their doors or windows, which means that the homes use more energy to keep the space comfortable.

-

The wood cook stove is the most used heating appliance and 162 people indicated when asked specifically, that this is due to the cost of the energy source. Households in these areas tend to consume more wood on average than the rural household in Kosovo with 10.35 solid m3 for low payment, 9.2 solid m3 for low-service and 8.2 solid m3 for the rural household (which was calculated from the UNFAO work)1.

-

The majority of all electric heating devices owned by the surveyed households are under five years old. Indicating that of all the household appliances, the heating sources are the most frequently replaced, when compared to the lifespan of other appliances found in the appliances section.

-

Electric heating devices are mostly used during winter nights for a couple of hours and were favored based on mobility. The electric quartz heater is the most used electric heating device because of its convenience and mobility

Appliances -

Water heaters consumed on average 212 kWh monthly. This was the highest consuming appliance, followed by the table top cookers (70.97 kWh), Televisions (tube 41 kWh, LCD 44 kWh) and vacuum cleaners (27 kWh).

UNFAO, “Wood Biomass Sector in Kosovo: Woodfuel Integrated Supply/Demand Overview Mapping (WISDOM).” 1

12/3/15


P.5

-

The age of electric appliances is contributing to the inefficient consumption of energy in households. Typical appliances had an average of a five year turnover rate. However, for water heaters (the most sued appliance in households) 24% (n=147) of the households surveyed had water heaters over ten years old.

-

Average wash machine use is 10.98 hours per week. Based on the information provided by households, on average, they launder 1.9 times the white clothes per week and 3.06 times the colored clothes per week. The average temperature for white laundry is 90 to 95 degrees, with only 3% (n=18) of the sample using colder temperatures. And the average temperature for colors is 40 -60 degrees Celsius

-

Water boilers have the option often to be turned off and on by a switch, therefore the average amount that the appliance is on is 8.93 hours during daytime and 9.57 during nighttime.

-

For showering purposes only, households consumed 212.5 kWh per month to heat the water. This amount was calculated based on the number of showers per person per week indicated on the survey and assumed the average hot water needed for Europe.

Customer Interaction -

KESCO teller windows are the customers’ main contact point, and could be used to reach target audiences for better understanding their bills or informing about efficiency. Other likely avenues to reach people are through mobile phones, TV and internet.

-

Twenty-seven percent of the sample always gets energy-related information through the TV, and research has shown that TV awareness campaigns are the most effective when shared by an impartial entity.

Isolation -

Half the sample had houses that were built after 2000. Older households tend to use more traditional forms of heating and less electricity.

-

Of the sample households, 17% (n=106) homes had both roof and wall insulation; 13% (n=80) had walls only and 7% (n=44) had roof only insulation. Of the sample, 44% of homes had a draft around their windows and doors. In terms of building materials, 448 of the households had double-pane windows.

-

The respondents with insulated walls and knew the dimensions, typically had insulation of 5cm (n:32), 8cm (n:22) or 10 cm (n:14); 14 residents indicated thicker

12/3/15


P.6

insulation than 10cm. The recommended width of insulation for households is 8-10 cm with appropriate sealing and spacing. Behaviors and Opinions -

Eighty-two percent of respondents indicated that they are concerned or very concerned about energy efficiency.

-

The 54% of respondents (n=327)tend to agree they are informed when it comes to energy efficiency. Another 18.5% (n=112) strongly agree that they are informed when it comes to efficiency.

-

Most of the respondents (83.3%, n=504) either strongly agree (44%, n=266) or tend to agree (39%, n=238) they would like to buy energy efficient appliances, but their financial situation makes it difficult.

-

Fifty-two percent (n=314) of households strongly agree and 36% (n=217) agree that electricity has become more expensive in the last 12 months.

-

The majority of respondents (88.5%, n=533) declared to never leave the lights on when they are not in the room. When crossing those that never leave their lights on when they’re not in the room averaged 4.62 – 4.64 hours for incandescent and CFL usage per day. Those (n=66) that said seldom or sometimes averaged 5.06-5.95 hours per day.

Investments -

When asked what investments a household will apply in the near future, a variety of options were asked. Wall insulation (57.7%, n=349) and then energy efficient appliances (73.7%, n=443) are the most likely investment that a household will make in the future

-

Most money for investments will most likely come from family or remittances to do these improvements. As those that had made improvements were asked where the funding came from, 66.8% (n=508) had used funds from their own or family resources, 21.8% (n=166) had used funds from remittances. Only 9.6% (n=73) indicated that they would take out a loan and 3 would take out a green loan. Multiple responses were allowed.

12/3/15


P.7

LIST OF ABBREVIATIONS CENR – RIT/AUK Center for Energy and Natural Resources KEDS – The Kosovo Energy Distribution and Supply Company (KEDS) is a joint stock company that operates throughout Kosovo and is the main electricity supply and distribution company. KESCO – Kosovo Electric Supply Company. KEK – The Kosovo Energy Corporation which produces energy throughout Kosovo LSLP – Low Service Low Payment areas UNFAO – United Nations Food And Agriculture Organization UNDP – United Nations Development Program

12/3/15


P.8

TABLE OF CONTENTS 1.

Introduction .................................................................................................................................................... 16

2.

Survey Methodology ................................................................................................................................... 17 2.1.

Survey Instrument .............................................................................................................................. 17

2.2.

Sample Selection .................................................................................................................................. 18

2.3.

Enumerator Training ......................................................................................................................... 19

2.4.

Data Collection ...................................................................................................................................... 20

2.4.1.

Demographics of the respondents ...................................................................................... 22

3.

Data Assumptions ........................................................................................................................................ 22

4.

Survey Results ............................................................................................................................................... 29 4.1.

Household Demographics ................................................................................................................ 29

4.1.1.

Demographics – family size and education ..................................................................... 29

4.1.2.

Employment – source of income, frequency, family businesses ............................ 32

4.1.3.

Remittances and Income ......................................................................................................... 36

4.2.

Energy Costs .......................................................................................................................................... 38

4.2.1.

Energy Prices ................................................................................................................................ 38

4.2.2.

Quantities of Alternative Energy Sources ........................................................................ 42

4.2.3.

Income and Expenditure on All Energy Sources ........................................................... 49

4.2.4.

Affordability .................................................................................................................................. 52

4.2.5.

Debt .................................................................................................................................................. 56

4.3.

KEDS and the Consumer ................................................................................................................... 59

4.3.1.

Bill Understanding and Electricity Consumption ......................................................... 59

4.3.2.

Electricity Price Variation Understanding and Electricity Consumption........... 61

4.3.3.

Improvements to the bill understanding ......................................................................... 63

4.3.4.

Communication ........................................................................................................................... 65

4.3.5.

Brand Outlook .............................................................................................................................. 76

4.4.

Household Construction ................................................................................................................... 78

4.4.1. 12/3/15

Age of Household and Surface Area.................................................................................... 78 Trends and Potentials for LPLS Households

P.9

4.4.2. 4.5.

Isolation .......................................................................................................................................... 83

Heating Systems ................................................................................................................................... 85

4.5.1.

Choice of Heating and Time of use ...................................................................................... 88

4.5.2.

Average Electricity Consumption for heating devices................................................ 91

4.6.

Appliances and Energy Consumption ......................................................................................... 92

4.6.1.

Water and Electricity ................................................................................................................ 92

4.6.2.

Household Appliances .............................................................................................................. 99

4.6.3.

Entertainment Appliances .................................................................................................... 111

4.6.4.

Computers ................................................................................................................................... 113

4.7.

Lighting .................................................................................................................................................. 115

4.8.

Behaviours and Energy Knowledge........................................................................................... 116

4.8.2.

Investments and Financing .................................................................................................. 120

5.

Recommendations ..................................................................................................................................... 122

6.

References ..................................................................................................................................................... 126

12/3/15


P.10

LIST OF FIGURES Figure 1. Difference in Energy Sources in Homes with Different Meters ......................................24 Figure 2. Household Electricity Bills Normalized by the Number of Household Members ......................................................................................................................................................26 Figure 3. The Number of Households by Income Category..................................................................33 Figure 4. Number of Households by Income Source and Debt ...........................................................34 Figure 5. Sources of Income and Average Winter kWh Consumption ............................................35 Figure 6. Households with Economic Activities Operating from the Home by Type of Meter .............................................................................................................................................................36 Figure 7. The Number of Households Receiving Remittances by the Frequency and Amount of Remittances .........................................................................................................................36 Figure 8. Feeling of Income Adequacy by Number of Household Members and Monthly Income Levels .........................................................................................................................37 Figure 9. Average Energy Expenditure per Month in Winter on different Energy Sources .........................................................................................................................................................40 Figure 10. Seasonal Prices of Wood Fuels ...................................................................................................41 Figure 11. Scatter Plot of the Share of Electricity for Digital Meter Consumer by Total Monthly Consumption (kWh).............................................................................................................43 Figure 12. Scatter Plot for Digital Meter Consumers for the Share of Electricity by the Share of Wood Fuels (kWh) ................................................................................................................43 Figure 13. Electricity Expenditure by Digital Meter Type, Income Level and Month (€) ...................................................................................................................................................................44 Figure 14. Electricity Expenditure by Analog Meter Type, Income Level and Month (€) ...................................................................................................................................................................45 Figure 15. Scatter Plot of the Share of Wood Fuels for Digital Meter Consumers by Total Monthly Consumption (kWh) .................................................................................................48 Figure 16. Scatter Plot of the Share of Lignite for Digital Meter Consumer by Total Monthly Consumption (kWh).............................................................................................................49 Figure 17. Energy Consumed by percent of income spent on energy .............................................50 Figure 18. Average Energy Consumed (kWh) by Income Category .................................................51 Figure 19. Average Monthly Expenditure by Income Level (€) .........................................................51 Figure 20. Percentage of Income Spent on Energy by Number of Households ...........................53 Figure 21. Social Assistance Metric by Income and Total People......................................................54 Figure 22. Households Qualifying for Social Assistance ........................................................................55

12/3/15


P.11

Figure 23. Energy Consumed by Households That Would Qualify for Social Assistance....................................................................................................................................................56 Figure 24. Income Spent on All Energy Resources and Whether or Not the Customer Has Debt .......................................................................................................................................................57 Figure 25. Income Levels and Debt.................................................................................................................57 Figure 26. Impact of Bill Understanding on Average Monthly Bills (€) for Digital Meters ...........................................................................................................................................................60 Figure 27. Impact of Bill Understanding on Average Monthly Bills (€) for Analog Meters ...........................................................................................................................................................61 Figure 28. Impact of Electricity Price Variation Understanding on Average Monthly Bills for Digital and Analog Meters...................................................................................................62 Figure 29. Location of Payment........................................................................................................................66 Figure 30. The Extent to which Media is followed for Information on Energy Efficiency or Outages ..............................................................................................................................67 Figure 31. The Extent to Which Media is followed for Information on Energy Efficiency or Outages based on Respondents' Age ....................................................................69 Figure 32. Households of Different Family Sizes Consulting the TV for Information on Energy Efficiency and Outages ....................................................................................................69 Figure 33. The Extent to which Media is followed for Information on Energy Efficiency or Outages by Respondents with Monthly Income under 300 euros ..........70 Figure 34. Households of Different Income Categories Consulting the TV to get information on Energy Efficiency and Outages ..........................................................................71 Figure 35. Education of Household Head and the Extent to which they consult the TV for Information on Energy Efficiency and Outages ...................................................................72 Figure 36. Quantity of Mobile Phones by household ..............................................................................74 Figure 37. Number of People who communicate with KESCO Employees when they come to Register Electricity Consumption ...................................................................................75 Figure 38. Change in household opinion before and after privatization .......................................78 Figure 39. Frequency by Age of House ..........................................................................................................80 Figure 40. Energy Consumption by Age of House ....................................................................................80 Figure 41. Surface Area by Age of House .....................................................................................................80 Figure 42. Surface Area and Space Heated ..................................................................................................80 Figure 43. Construction Material for the walls ..........................................................................................81 Figure 44. Construction Material for the Roof ...........................................................................................81 Figure 45. Construction Material for the Flooring ...................................................................................82

12/3/15


P.12

Figure 46. Draft and Window Materials .......................................................................................................83 Figure 47. Insulation by age of building .......................................................................................................83 Figure 48. Insulation by size of building and energy expenditure....................................................84 Figure 49. Consumption of kWh by Insulation and space heated (m2) of building ...................85 Figure 50. Heating Systems Ranked by Usage ...........................................................................................87 Figure 51. Reasons behind Why Respondents Use their Top Three Heating Systems ............88 Figure 52. Hours an Electric Heating Device is used...............................................................................89 Figure 53. Ages of Electric Heating Devices ................................................................................................91 Figure 54. Size of Water Boilers in kW .........................................................................................................95 Figure 55. Size of Water Boilers in liters ......................................................................................................96 Figure 56. Age of Utility Appliances ............................................................................................................ 100 Figure 57. Vacuum cleaner usage ................................................................................................................. 101 Figure 58. Laundering Frequency ................................................................................................................ 103 Figure 59. Washing Machine Usage and Frequency of Loads .......................................................... 104 Figure 60. Age of Large Kitchen Appliances ............................................................................................ 106 Figure 61. Usage of Electric Cooking Devices.......................................................................................... 108 Figure 62. Age of Cooking Appliances ........................................................................................................ 109 Figure 63 Usage of TV, LCD/Plasmas, Digital Receivers .................................................................... 112 Figure 64 Age of TV, LCD/Plasmas, Digital Receivers ......................................................................... 112 Figure 65 Age of Laptops and Desktop Computers .............................................................................. 114 Figure 66 Usage of Desktops (PC) and Laptops ..................................................................................... 115 Figure 67. Use of Incandescent and CFL Lights in Hours ................................................................... 116 Figure 68. Respondents' Concern about Energy Efficiency .............................................................. 117 Figure 69. Respondents' Agreement or Disagreement with Behavior-Related Statements ............................................................................................................................................... 119 Figure 70. Water and Heat Conservation Practices of Respondents ............................................ 120 Figure 71. Likelihood of applying different energy saving measures .......................................... 121 Figure 72. Sources of financing for energy efficient improvements ............................................. 121

12/3/15


P.13

LIST OF TABLES

Table 1. Breakdown of Survey Instrument .................................................................................................17 Table 2. Surveys Completed in the Municipalities and Village (N: 605) ........................................21 Table 3. Age and Gender of Survey Respondents .....................................................................................22 Table 4. Tariff Rates for Analog and Digital Meters, by Season and Time (euro cents/kWh).................................................................................................................................................23 Table 5. Descriptive Statistics of Electricity Bills (€)..............................................................................25 Table 6. Average Per Person Monthly Consumption for the Digital Meters (€) .........................26 Table 7. Average Per Person Monthly Consumption for the Analog Meters (€) ........................26 Table 8. Conversion Factors for Energy Sources ......................................................................................28 Table 9. Distribution of Surveyed Households by Tariff Block and Number of Members ......................................................................................................................................................30 Table 10. Households by Members (n: 605) ...............................................................................................31 Table 11. Education Attainment by Position in the Household .........................................................32 Table 12. Education Attainment and Average Energy Consumption ..............................................32 Table 13. Household Employment by Gender (n: 605) .........................................................................33 Table 14. Household Income Adequacy and Debt Distribution .........................................................38 Table 15. Average Energy Expenditure per Month in winter on Different Energy Sources .........................................................................................................................................................39 Table 16. Prices for Energy Sources 2014/15 ...........................................................................................39 Table 17. Average Annual Wood Fuel Consumption for Kosovo .......................................................41 Table 18. Descriptive Statistics of Wood fuel Consumption ................................................................47 Table 19. Descriptive Statistics of Non-Electricity kWh Consumed .................................................50 Table 20. Consumption Difference for by Social Assistance Metric (kWh, digital Meters) .........................................................................................................................................................55 Table 21. Communication and Bill Understanding Variables .............................................................76 Table 22. Relationship between KEDS Communication and Customers' rating of their Services before and after Privatization (N:590) .............................................................77 Table 23. Number of households and their opinions on the distribution pre- and postprivatization ...................................................................................................................................78 Table 24. Combination of insulation, windows and drafts ...................................................................84 Table 25. Electric Heating Devices Average Electricity Consumed ..................................................91 Table 26. Average kWh consumption for appliances .............................................................................92 Table 27. Age of Water Boilers (N: 555) .......................................................................................................94 12/3/15


P.14

Table 28. Water Heaters kW and Liter Range............................................................................................94 Table 29. Hours Used for Water Boilers .......................................................................................................94 Table 30. Specifications of water boilers .....................................................................................................96 Table 31. Average amount of kWh used, based on the number of people and the number of showers each week ..........................................................................................................98 Table 32. Turnover rate for Utility Appliances by Year of Home Construction (n=605) ..................................................................................................................................................... 101 Table 33 Age of Washing Machines ............................................................................................................. 103 Table 34 Brand of Washing Machines ........................................................................................................ 104 Table 35. Turnover rate for Cooling Appliances by Year of Home Construction (n=605) ..................................................................................................................................................... 106 Table 36 Brand of Fridge-freezers ............................................................................................................... 107 Table 37. Turnover rate for Cooking Appliances by Year of Home Construction (n=605) ..................................................................................................................................................... 111

12/3/15


P.15

This document is to report on the findings of the RIT Center for Energy and Natural Resources (CENR) household survey covering energy related factors for households in areas that were identified as low-service and low-payment throughout Kosovo. This document reports on the research methodology, survey design, data collection and research findings. This survey will be referenced in the rest of the document as the 2015 Energy Survey and the data as the low-payment, or low-service quality (LPLS) dataset. The data from the survey will be reported with relevant statistics on consumption in order to create a clear picture of household characteristics, energy consumption, home building materials, appliances, behaviors and well-being. These statistics are consistent to those found in the literature and some are expandable to general statistics collected in Kosovo. 1. INTRODUCTION The energy profile of households in Kosovo has four main components that contribute to electric demand – heating, cooking, appliances, lighting. Managing these requires developing an understanding of the underlying behavioral and economic aspects that drive electric use. Often, as costs of use increase, household decision makers will switch consumption patterns; they will attempt to conserve or change energy sources. As developing countries move toward modern forms of energy and away from traditional fuels (e.g. wood, biomass) there are likely to be decisions based on reliability and availability of other energy sources as well as the cost aspects that drive these choices. Most areas that were surveyed in Kosovo are considered rural, except in the select neighborhoods of Prizren, Prishtina, Podujevo and Ferizaj. Rural households differ in energy consumption patterns, especially in developing countries, by the higher usage of wood fuels, relative price differences between energy sources, differences in income sources and also cultural practices relating to heating and cooking2. It is also assumed that as incomes and prices change that rural households will be less likely to fully transition away from traditional wood fuels, and instead accumulate their energy options3. Therefore, changing an aspect of the energy system may not have the immediate or desired outcome. The households that were surveyed for this analysis were designated in areas of either low-payment or low-service quality by the energy distribution company KEDS. The objectives of the report are to focus on the interactions of these low-income and lowKowsari and Zerriffi, “Three Dimensional Energy Profile:: A Conceptual Framework for Assessing Household Energy Use.” 3 Masera, Saatkamp, and Kammen, “From Linear Fuel Switching to Multiple Cooking Strategies.” 2

12/3/15


P.16

service area households and their interactions with the energy system, focusing on how they are consuming electricity, reasons for low-payment and areas for intervention strategies. Characteristics of the residents, their living environment, their appliances and activities, behaviors and knowledge and the subsequent effects on the energy consumption are specifically described. Additionally, customer interactions with the energy distribution company, KEDS are analyzed. With the data collected, there are several important policy and business related questions that can be answered, especially in regards to affordability, energy efficiency and customer relations. This report is organized to cover the survey methodology, survey results, a review of similar studies and policy recommendations throughout the different sections. 2. SURVEY METHODOLOGY 2.1.

Survey Instrument

The objective of the survey is to collect useful end-user data in order to make informed decisions for policy and investments in households. In order to build an appropriate survey for household energy consumption, several surveys were consulted along with the Energy Efficiency Indicators from the International Energy Agency4. The survey instrument covers the following sections of energy consumption in the household (Table 1). The survey took on average 33 minutes, with the longest survey taking 1:40, which would happen when survey respondents would let the enumerators into their home to look at appliances. Table 1. Breakdown of Survey Instrument Section Household Characteristics Building Characteristics Heating and Cooling Electricity and Water Characteristics and Use of Appliances Behavior Patterns Follow-up

Number of Questions 12 14 17 9 6 6 6

The instrument went through a variety of versions and was pre-tested in 60 households in the village of Havajli as part of a senior capstone project by former CENR employee Dina Vllasaliu. The pre-testing was critical, as the understanding of words like 4

IEA, “Energy Efficiency Indicators: Fundamentals on Statistics.”

12/3/15


P.17

‘efficiency’ which has no translation in either Albanian or Serbian were tested and altered to get the understanding that energy efficiency means energy savings to the population. Any question that got an ambiguous response was corrected and then retested. The final version of the survey was translated into Albanian and Serbian and was accompanied by a set of pictures relating to the building materials and energy appliances. Households were told at the beginning of the survey that their participation was voluntary and that they could refuse to answer any question that they were not comfortable answering. These responses were coded as, Not Applicable (777); ‘I do not know’ (888) and ‘Do not want to respond’ (999), in order to make sure that non-responses were not coded as zeroes, when appropriate. Additionally, respondents could say ‘It depends’ (4444) and ‘Rarely’ (3333). The questions regarding personally identifying information (name and contact info) were separated and excluded from the final dataset for confidentiality. Additionally, customer numbers were not asked of survey participants as the PI felt that it would be a breach of confidentiality to connect households to their specific KEDS records. 2.2.

Sample Selection

The specific sample for this surveying effort was to analyze households within areas that had either low payment, or low service quality (LPLS). This stratification was identified as two priority areas from KEDS and the list of villages were preselected. These households were selected in order to better understand how to meet the electric needs of the consumers while minimizing the costs of supplying these households and to better understand how to minimize disconnections. A sample size of 605 was chosen for this survey, which represents 4.8% of the customers in the selected population5. In order to make the sample robust to other statistics collected in the country, the sample was stratified in two ways. The first stratification took the population of households from the Census for each of the municipalities and the total households for the rural areas and created a weight6. The population weight was applied to the total number of surveys to get the number of surveys needed by municipality to be representative of the total population. Since not all municipalities were covered, the remainder of surveys was then weighted by the number of customers in low-service and low payment areas (since there may be more in some 5 6

N=12,604 that are either low payment or low service areas KSA, “Kosovo Population and Housing Census 2011: Final Results.”

12/3/15


P.18

municipalities than others). This total was then divided by the number of LPLS villages in each municipality to get a minimum needed from each village. The second stratification was based on electric substations; each of the villages in the substation a weight was created based on the number of customers. If there was only one village for the substation, the village received the minimum surveys needed for their municipality. Most substations had multiple villages (e.g. the Bellanice substation has 4 villages). The villages were selected randomly by random number generation, based on the number of customers in the substation and village, the final number of surveys per village was based on the second stratification. For example Ferizaj has 10,575 rural households, in Kosovo there are a total of 168,959 rural households; therefore Ferizaj has 6% of the rural households so 6% of the total surveys needed to come from Feraizaj. Of the surveys remaining, Ferizaj had 4% of the LSLP customers, therefore an additional surveys were allocated to Ferizaj, for a total of 31 surveys needed. There were 4 LPLS villages in Ferizaj but only 3 substations, the 2 villages in the same substation were weighted based on the number of customers in that substation in relation to the rest of the Ferizaj stations. The random number generator selected the villages of Bablak, Sherret (Ferizaj) and Trimor, which were the only villages for their respective substations, so the number of surveys from Ferizaj was split between the three villages based on the number of customers in each. Of the total households surveyed, the survey equally split between the low-service and low-payment areas, with 292 of the households in each. The remaining 21 households were from villages randomly selected and not within the original lists of villages. 2.3.

Enumerator Training

The enumerators for the survey were a combination of University of Pristina students enrolled in the KEDS Academy and Research Assistants from the CENR. Each person that went into the villages and interacted with the survey participants also took human subjects training, as to understand that the data that the households were providing can be sensitive in nature and no household should be prodded into answering any question. The students were trained and practiced giving the surveys prior to going into the villages. The teams were additionally monitored in the field by CENR staff in order to make sure that the questions were being asked in accordance to the written questions with as little ambiguity

12/3/15


P.19

possible. Most days there were 5 to 6 groups visiting the villages, paired off male and female, for cultural reasons to visit the households. 2.4.

Data Collection

A total of 40 villages were visited in 17 municipalities throughout Kosovo from April 12, 2015 to May 7, 2015. On average, there were 5 to 6 teams of two students surveying in a village. In most of the villages, the main square, mosque or school was chosen as the starting point and the enumerators expanded outwards from the center, randomly selecting household participants. Some of the villages that were selected had issues in terms of survey collection. For example, when surveying Bablak, most households were empty as it was a relocation village. In Peja, the villages of Raushiq/Novoselle and Zllapek had a funeral on the day of the visit, so fewer households were able to be surveyed. An additional village of Brezhanik was selected and visited on a different day, in order to get more households from this municipality. Additionally, the village of Pllaqice was not on our original village selection, as it was unmarked and shared the same school as Marali, but given the proximity to the village, the households were still included in the sample. The resulting villages sampled for the survey can be found in Table 2.

12/3/15


P.20

Table 2. Surveys Completed in the Municipalities and Village (N: 605) Municipality Decan Drenas Ferizaj Fushe Kosove Gjakove Gjakove Kaqanik Klina Lipjan Lipjan Malisheve

Malisheve

12/3/15

Village Carabreg Voksh Arllat Krotice e ulet Bablak Sherret Trim Vragoli Bec Cermjan Ramoc Nikaj Caravike Baince Gadime Kraishte Magure Qylage Bellanice Lladrovc Marali Pllaqice

N 6 8 11 14 10 8 13 30 21 24 5 24 6 11 11 28 11 10 21 13 5 14

Senik

19

Municipality Mitrovice Obiliq Peja

Podejeve Prishtina Prizren Rahovec Shtime Suharek Viti Vushtrri

Village Pirq Verrnice Lajthishte Brezhanik Novoselle Zllapek Blakshort Llapashtice Podejeve Shakofce Vashariste Cameri Atmaxha Hoqe e Qytetit Lubishte Krushe e Madre Radoste Recak Greikoc Germova Gushice Bukosh Kolle Oshlan


N 11 10 10 11 7 8 4 12 5 12 9 10 16 19 11 9 10 11 19 24 20 14 9 11

P.21

2.4.1.

Demographics of the respondents

In order to understand the data, the following section describes the characteristics of the survey respondents. As part of the human subject guidelines, no survey respondent was to be interviewed under the age of 18. The oldest survey participant was 85 years old. The gender balance of the survey respondents was dominantly male (62.8%), though for household cleaning activities, often a female household member was also interviewed. In several households, multiple household members were participating in the answering of questions. Households were interviewed throughout the day and therefore a variety of different individuals were available. The sample of households and villages were predominantly Albanian, with 8% (n=49) of the survey respondents from the minority communities of Bosnian, Gorani and Roma. The lack of Serbian households is a limitation of the survey. Table 3. Age and Gender of Survey Respondents Age 18 – 30 31- 40 41 – 50 51 – 60 60+ Average

Gender Male Female 77 60 68 46 72 73 72 20 90 19 46.55 40.7

Of the members primarily interviewed, 52.6 percent were the head of household/main decision maker. Although, culturally the head of house typically makes many of the logistical decisions, they may not necessarily be the main income earner. Of the 605 households interviewed, 97 survey participants were the head of house, but not the main income earner. Conversely, 22 members were not the head of household but indicated that they were the main income earner. 3. DATA ASSUMPTIONS Based on the data collected more calculations need to be made with regards to energy consumption. The following are the assumptions made in regards to calculating the final numbers.

12/3/15


P.22

KEDS has undertaken the action of replacing analog electricity meters with digital ones. Since these systems are new and may have altered consumption or understanding, the analysis will distinguish between households that have digital and analog meters. Of the 605 households surveyed, 320 had the new digital meters and 277 had the analog meters (8 households did not know or did not respond). All of the digital meters are on the two tariff (high/low) three-block structure, which changes seasonally. The analog meters are a combination of the two- and one-tariff (no high/low) block structures. The survey did not directly ask the customer numbers of the households and thus cannot separate the analog meter household bills into the different consumption. There is a significant difference in how these monthly bills would be back-calculated to kWh consumption (Table 4). Additionally to calculate a household bill a €2.5 fixed charge is applied and then a VAT of 16% is added, this was adjusted from the bill amounts that the households provided. Table 4. Tariff Rates for Analog and Digital Meters, by Season and Time (euro cents/kWh) 2014

Block 600 Source:

Day/Night high low high low high low Average

0.4kV Domestic 2-rate meter Oct- Mar April – Sept 5.28 3.8 2.66 1.89 7.32 5.24 3.67 2.62 10.62 7.6 5.3 3.82 5.81 4.16

0.4kV Domestic 1- rate meter Oct- Mar April - Sept 4.72

3.37

6.53

4.66

9.46

6.79

6.90

4.94

7

The amounts were adjusted by the fixed fee and the 16% VAT on electricity and then the consumption for the different energy blocks given the tariff prices for 2014 were then solved for using Microsoft Excel’s solver function. Therefore, when backing out the ranges of expenditure, for the 2014 2-rate meters first tariff block (0 -200 kWh) will have a bill less than €15.15 or 13.36 if it follows the expected 70/30 day/night split used by the ERO8. For households consuming in the second tariff block (200-400 kWh), the upper-end of their monthly consumption would be €49.11 to €42.21, respectively. For the bills

7 8

ERO, 2014 Tariff Rates for Kosovo. ERO, “Consultation Paper on Tariff Structure.”

12/3/15


P.23

between these ranges, the charge will depend on how many day hours and how many night hours are used, hence why the distribution of prices matters. For the single tariff meters, spending under 200 kWh would result in a maximum bill of €13.85, for the second tariff block the maximum would be €49.26. Day and night consumption for the single tariff block decreases the distributional gains by consuming at night at a lower price. Therefore, in the analysis the statistics relating to monthly bills and their respective kWh consumption is split into the two separate categories. Given the difference in prices for analog meters (and some question of the reliability actual consumption), when calculating the kWh for the analysis within the report for kWh comparisons, it only include households with digital meters. The amounts for the different low and high tariffs were summed by block to get a total kWh consumed in a month. The other factors of energy fuel consumption were not significantly different between the analog and digital meters (Figure 1), therefore the estimates for price and quantity of energy will be based on the digital meters. The assumptions for the calculation of the kWh equivalents are explained later in this section.

Monthly kWh

1,400 1,200

Average Winter Electricity

1,000

Solid Wood Adjusted

800 Lignite

600

Fuel Oil

400 200

Natural Gas

0 Digital (new) Analog (old)

Figure 1. Difference in Energy Sources in Homes with Different Meters Since the monthly bills will be used throughout the analysis the general descriptive statistics will be presented here. Consumers were also asked what the amounts of their bills were for the months of July 2014 and October through March 2015. Half way through the survey, the customers had received their April bills and were willing to share that information as well. Most people recalled the information from memory; however, 63

12/3/15


P.24

households gave us at least one paper bill. The descriptive statistics for the different months of bills are found in Table 5. Since the amount of energy that is spent in a household is dependent on the number of people living in the home, the bills were adjusted to the average amounts spent per person. Table 5. Descriptive Statistics of Electricity Bills (€)

July October November December January February March April Average Winter Bill Average Summer Bills

Num. Hhhld 425 466 473 498 541 546 567 116 584 497

Minimum

Maximum

Mean

2.90 2.90 2.90 2.90 2.90 2.50 2.50 2.50 3.77 2.90

190 280 320 300 280 390 312 300 295 190

29.55 38.52 41.68 46.28 49.50 42.77 40.35 34.08 43.30 30.01

Std. Deviation 18.00 23.96 26.02 27.52 30.03 28.22 28.36 30.33 26.19 18.87

Considering the number of people in the household and the monthly electricity bills, the per-person average is approximately €7.2 per person for the new digital meters and €6.7 per person for the old meters (Table 6 and Table 7). The distribution of people and the expenditures on electricity are also important, as households with fewer people may be using energy less efficiently than households with more individuals. From Figure 2, as the number of household members decrease, household consumption tends to level out at around 5 to 6 people, when the marginal contribution of the additional person is initially negative. Increasing the number of people past this amount, fluctuates the impact on the monthly bills.

12/3/15


P.25

Table 6. Average Per Person Monthly Consumption for the Digital Meters (€)

Mean July October November December January February March April

5.33 6.65 7.24 8.09 8.88 7.50 6.93 6.98

Digital (new) Std. Minimum Deviation 3.89 0.48 4.49 0.48 4.87 0.48 5.37 0.48 6.79 0.48 5.27 0.42 5.31 0.42 5.97 0.50

Maximum 25.00 34.00 34.00 38.00 55.00 36.00 62.40 42.86

Table 7. Average Per Person Monthly Consumption for the Analog Meters (€) Analog (old) Mean July October November December January February March April

4.75 6.50 7.06 7.68 8.11 7.21 6.63 5.61

Std. Deviation 2.39 4.50 4.95 4.90 4.96 5.40 4.02 3.44

Minimum

Maximum

0.83 0.92 0.42 0.85 0.53 0.94 1.25 2.00

15.00 46.67 53.33 50.00 46.67 65.00 33.33 19.00

Figure 2. Household Electricity Bills Normalized by the Number of Household Members

12/3/15


P.26

Households were asked about other sources of energy that they consume, including the total cost, unit prices and quantities. When considering the amount of money spent non-electric sources of energy, the amount spent per month for wood fuels was decreased to adjust for households that use wood year round. The average quantity for the sample for using wood fuels ‘only during the heating season’ was 8.47 solid m3 and the mean for ‘all year round’ was 10.46 solid m3. Therefore to estimate only the winter heating for these houses, the quantity was adjusted by 2.08, and total cost was decreased by the price paid multiplied by 2.08. Some households harvested from their own forests and indicated a price of zero for the wood, opportunity costs were not calculated for these households. For the other energy types consumed, coal, natural gas and fuel oil, the total spent was divided by 7 to compare to the monthly amount spent for electricity since these fuel sources were asked about the seven month heating season. In order to compare the energy content and prices of the different sources, the quantities of the different fuels and their prices were multiplied using the conversion factors their sources in Table 8. There were assumptions made for wood fuels based on the timing of wood harvested relative to the energy season. The energy content assumed for wood fuels is based on ‘green’ oak with 68% moisture content. Ideally if wood was allowed to season for a year, the kWh content of it would be closer to 1,400 – 2,000 kWh/m3. The energy potential of lignite is based on numbers from previous work with KOSTT on the caloric values of lignite provided to Kosovo A and B.

12/3/15


P.27

Table 8. Conversion Factors for Energy Sources Unit 12 MWh 42 Gj of heat 1 cubic meter fuel wood; 1 metric ton fuel wood From Solid cubic Meters to kWh 1.075 tons briquettes 1 m3 of Briquettes From Briquette to kWh 1 ton of Pellets 1 ton lignite Coal 1 litre Heating Oil I litre Heavy Fuel Oil I litre Natural Gas 1 m3 of Natural Gas

= = = = = = = = = = = = = =

Conversion 1 toe 1 toe 0.3531 metric ton 0.3215 tons of oil equivalent (toe) Multiply by a factor of 734.07 m3 0.09 gj Multiply by a factor of 20.507 4.85 MWh 70% of 1.83 MWh 10 kWh 11.45 kWh 3.71 m3 6.6 kWh

Source 9 10 11 12

13 14

15 16 17 18 19 20

BP, “Conversion Factors.” Ibid. 11 UNECE, “Handy Guide to Wood Energy.” 12 Ibid. 13 UNECE, “Forest Product Conversion Factors: Project Overview and Status.” 14 Ibid. 15 Ibid. 16 KOSTT, “Measurements for the Lignite Power Plants A, B and Proposed”; “How to Measure Fuel Efficiency, Energy Costs, and Carbon Emissions for Home Heating.”; assuming a 70% efficiency in the production at home in stoves vs, the caloric content on average of lignite in Kosovo 17 SEAI, “Comparison of Energy Costs Commercial/Industrial Fuels.” 18 Ibid. 19 Ibid. 20 Ibid. 9

10

12/3/15


P.28

4. SURVEY RESULTS The results from the household survey can be found in the following sections for the 605 households that were surveyed throughout Kosovo. Within each section there are results from the survey, recommendations based on the data analysis and relevant literature available. In general, when considering all of the factors that affect household electricity consumption for the average of the winter monthly bills, the household factors collected were correlated and run through a predictive analysis model in the SPSS statistics package. The top factor in predicting the total amount spent for electricity was where the main source of income for the family was a ‘family or owned business’ (14%). The total number of people in the home (12%), firewood consumption (8%), area of heated rooms (7%), village (6%), income (6%), TV usage (6%) and the rating regarding the quality of KEDS Services (5%) were additionally calculated as predictors of the electricity bills. These factors will be closely considered in the following analysis. 4.1.

Household Demographics

The characteristics in terms of the determinants of energy consumption in the home are the number of household members, income and education. The number of people in a household has been linked to the changes in fuel sources and increased consumption of energy, in developing countries this has been marked by a reliance on wood fuels21. In several countries, the number of people is a more significant determinant of energy consumption than income22. Though not to minimize income, which is key to the household being able to afford to increase the quantity and quality of energy sources. The average size of Kosovar families and income levels these factors are key to understanding the energy consumption of the home. The following section covers demographics of the households surveyed. 4.1.1.

Demographics – family size and education Households were asked about the structure of the people living within the home.

The responses were limited to a core family, consisting of the grandparents, parents and children; or an extended family that included spouses and non-direct relatives in addition to the core family. It was asked in this way, because it is typical of Albanian households to Link, Axinn, and Ghimire, “Household Energy Consumption: Community Context and the Fuelwood Transition.” 22 FAO, “The Role of Wood Energy in Asia.” 21

12/3/15


P.29

live with multiple relatives and extended families. This increases the amount of net energy needed, but also decreases the marginal cost of providing that energy. The respondents that indicated that only the core family was living in the residence (n: 531), larger extended families (n: 63) were common amongst respondents over the age of 50. The larger the family demand for electricity, the faster the household may increase to larger tariff block (Table 9). Using the family demographics and average bill consumption in the tariff blocks, the smaller the family the more likely they were to increase the expenditure. Households over a certain size or those with multiple families may benefit from dividing the household into multiple paying units to decrease the premiums paid for consuming over 600 kWh monthly. Using a shorthanded calculation, this would have to be weighed with the marginal cost of the additional person in the household (approx. € 7.2 for over 6 people) and the fixed cost of a new meter (€2.5) plus the consumption charges per additional person (the difference of €4.7 amounts to approximately 115 kWh). This would go against the idea of conservation of energy resources, but it may make electricity more affordable for larger families. Table 9. Distribution of Surveyed Households by Tariff Block and Number of Members

Tariff blocks

600

Number of Household Members 1-5 6- 10 > 10 14 11 3 98 65 15 20 71 22

From Table 10, the average number of people living within the home is 7 people, though larger families were not uncommon, with 79 households having more than 10 people (not including diaspora and students living abroad). The most frequent household sizes were 5 to 6 people. In terms of the gender balance of the overall household are evenly split with the average home being half female. Typically, there are 2 people on average under 18 and 433 have at least one dependent. Households on average have 1.5 members living outside of the home (diaspora and students), but the frequency of homes with at least one person in this category is only 47% (n: 286) of the sample indicating that these households may not have external help. The maximum any one family had in the diaspora was 22 members. The number of diaspora has implications for the availability of funds outside of the home. Additionally it has been found that households with diaspora

12/3/15


P.30

members typically have more electronic devices than households that do not23. Households in both categories of low-payment and low service had an average of 1.88 and 1.38 persons within the diaspora. There were 317 households that had no member living outside of the home. Table 10. Households by Members (n: 605) Mean Total People Male Female Under 18 Living in the household for less than 6 months (diaspora and students)

6.52 3.23 3.30 1.91 1.65

Std. Maximum Deviation 3.05 25 1.78 15 1.85 12 1.89 12 2.87

22

Given that on average half of the household is women, there is not a statistically significant correlation between electricity consumption and gender when normalized to the number of people living in the home. When tested separately each additional female in the household contributed on average 29 kWh, whereas an additional man contributed on average 45 kWh to the monthly winter bill. As the number of people living in the household for less than 6 months increases, it contributes to 12 kWh of additional consumption. From the literature24, as education increases, the tendency is to decrease the amount of energy that is being consumed. In terms of education, most households are only educated to the level of high school (Table 11). This will have a likely impact on household earning potential and the potential to understand the complexities of household energy consumption and energy efficiency. Although it is likely that they understand the specific economic tradeoffs of consumption, these challenges will need to be considered when designing programs around education and investment – in a sense meeting the consumer at their level of education. Energy consumption of households by education level can be found in Table 11, with the data for the singular households data omitted. As the head of household’s education increases, the reverse is true, with more energy consumed with the higher levels of education. UNDP, “Kosovo Human Development Report 2014.” Nikodinoska, “Determinants and Development of Electricity Consumption of German Households over Time”; Kristiansen, “Does Information Lead to Household Electricity Conservation?”; Abdelkarim et al., “Testing the Causality between Electricity Consumption, Energy Use and Education in Africa.” 23 24

12/3/15


P.31

Table 11. Education Attainment by Position in the Household Head of House None Elementary High School Undergraduate Graduate Other

1 135 211 60 0 0

Survey Respondent not Head of House 1 94 125 44 4 1

Main Earner, not Head of House 0 1 12 8 1 0

Note: Some survey respondents also answered the education attainment for the head of household when it was not them.

Table 12. Education Attainment and Average Energy Consumption Respondent’s (Not Head of House) Education Average Average Average Average Monthly Winter Monthly Winter Consumption Electricity Consumption Electricity (only ekWh) ( kWh) (only ekWh) ( kWh) 1,147.88 487.42 1,210.73 499.98 1,147.04 518.76 1,295.77 525.76 1,188.59 556.65 1,229.36 728.65 1,058.65 405.80 Head of House's Education

Education Level

None Elementary High School Undergraduate Graduate Other

4.1.2.

Employment – source of income, frequency, family businesses

Income is also important to understand within the sample population. There are different avenues in which households can support themselves financially – employment, diaspora and social assistance (pensions, etc.). In 36.5% (n=220) of all households in which no person is working, regardless of gender. Of these households that have no members employed, 116 households also do not have any people in the diaspora (Table 13).

12/3/15


P.32

Table 13. Household Employment by Gender (n: 605)

Number of people working 0 1 2 3 4 5 6 14

Number of Households Male Female 228 538 257 55 89 5 23 3 4 0 1 0 1 0 1 0

In terms of monthly income for the household, the most frequent income category for households was a monthly income between €200 and €300 (Figure 3). For a household in the average income category to not be considered energy poor, the total amount spent monthly would need to be in the range of €20 to €30. Though, the 10% rule common in energy studies may not be the most effective measure of energy poverty in Kosovo. This would indicate that energy poverty extends into the 200-600 (approximately 540 kWh) tariff block and the costs wood fuels would not be included.

120 100 80 60 40 20 0

Figure 3. The Number of Households by Income Category

12/3/15


P.33

The financial opportunities in households come from a variety of sources. Income sources have an impact on the ability to pay monthly bills, with agriculture and remittances as infrequent sources, and pensioners being less able to adjust in price increases without giving up consumption in other areas if possible. When asked about the main sources of incomes for the household, full-time jobs, pensions, agriculture/ livestock and remittances were all important sources. In agriculture, there were 153 households with at least one member, 13 with two members and 18 with 3 or more people working in the sector. Remittances were similar, with 129 households with at least one member giving assistance, 26 with 2 household members and 23 households with 3 or more. Homes with remittances had less frequency of debts than those in agriculture (Figure 4).

Other

Domesic Financial Help

Full-time Job 350 300 250 200 150 100 50 0

Remittances

Part-Time job

Family Business Total Debt Pension

Agriculture/livesto ck

Social Assistance Unemployment

Figure 4. Number of Households by Income Source and Debt Households may be constrained by their monthly income and therefore likely decrease the amount of electricity that they consume. Looking at the houses with digital meters there are several interesting groupings to consider (Figure 5). Households that are unemployed have a higher percentage consuming less than 200 kWh monthly. Those that have social assistance, tend to have a higher percentage consuming in the second tariff block, as holds with those that have part time jobs and receive domestic assistance.

12/3/15


P.34

% of HHLDS in the category

600

100% 90% 80% 70% 60% 50% 40% 30% 20% 10% 0%

Source of Income

Figure 5. Sources of Income and Average Winter kWh Consumption Family businesses account for 73 household’s main source of income, often with multiple family members working in the same business. When households were asked if there is any business operating on the household’s electric meter, twenty-nine households said that they were using their household meters for economic activities. These households had bills that were sizably higher than the general sample for both digital and analog meters (Figure 6). Households are subsidizing their economic activities by €13 to €30 per month. More information would be needed to ascertain whether or not these businesses

Bill Amount (Euro)

are sizable with storefronts or more along the lines of handicraft/microenterprises. 80.00 70.00 60.00 50.00 40.00 30.00 20.00 10.00 -

Jul

Oct

Nov

Dec

Jan

Feb

Mar

No Business | Digital

30.10

38.35

41.57

46.66

49.71

42.31

40.29

Business | Digital

56.18

62.68

66.91

69.44

73.76

68.68

69.41

No Business | Analog

26.50

35.95

38.73

42.51

46.04

40.41

37.08

Business |Analog

44.75

49.33

59.78

72.64

72.04

61.53

65.00

12/3/15


P.35

Figure 6. Households with Economic Activities Operating from the Home by Type of Meter 4.1.3.

Remittances and Income Households were asked what the monthly income was for the household and what

the net remittances were for the year 2014 and the frequency of receiving them. The number of households that are not receiving any remittances was 54% (n=326) of the sample. Most households that did receive remittances, were ‘as needed,’ meaning that it’s

Number of Homes

unlikely that they could be counted as a stable source of income for a household (Figure 7). 100 90 80 70 60 50 40 30 20 10 0

Other None As needed Yearly Semi-annual Monthly Weekly

Income Level

Figure 7. The Number of Households Receiving Remittances by the Frequency and Amount of Remittances Households were asked if they felt their incomes satisfies their basic needs; the responses were scaled as “Always, Often, Sometimes, Seldom, Never, and Don’t Know.” Although this measure is subjective on its own, combined with the response on the number of people within the household and monthly income, the measure may give some indications for relative poverty and directly correlates with their energy poverty (Figure 8). Respondents that feel like their monthly incomes always satisfy their needs tend to be smaller families and make closer to €600-700 per month. As the feeling of adequacy decreases, the level of income decreases and the number of residents increase in variation. Respondents that indicated that their incomes ‘Never’ satisfy their basic needs had incomes between €100- 200 per month and had six household members.

12/3/15


P.36

Figure 8. Feeling of Income Adequacy by Number of Household Members and Monthly Income Levels When considering income adequacy and the incidences of debt, only the category of ‘Never’ had higher respondents with debts (Table 14). This indicates that aside from the very income poor, households are no more likely based on the adequacy of their income to carry debt. More information on the factors that correlate with households having debts will be covered in a following section.

12/3/15


P.37

Table 14. Household Income Adequacy and Debt Distribution Do you have an outstanding balance with either KEK/KEDS/KESCO?

Do you feel like your monthly income satisfies your basic needs?

Always Often Sometimes Seldom Never

No

Yes

33 47 60 53 117

22 23 55 49 128

4.2. Energy Costs 4.2.1.

Energy Prices

Economically, the subsidization of energy resources may have two different outcomes – a substitution effect and a consumption effect. As the relative prices between energy sources changes, people will shift consumption to the lowest cost source. If the subsidization is not specific to any one energy source in terms of price or quantity, then households may consume more of all energy sources in some proportion or consume other goods (assuming that their energy needs are satiated). Additionally, the price of alternative goods is expected to fluctuate prior and during the winter months, based on demand. For the LPLS data, the most frequently consumed source was wood fuels, followed by natural gas, and lignite (Table 16). Lignite prices and wood fuels during certain parts of the season were equal per unit. However, when comparing them in terms of their equivalent kWh prices25, depending on when the timing of use and position in the monthly tariff block, the kWh equivalent of many sources is cheaper than electricity. The relative price difference should be considered when considering an increase the electricity share of the total energy budget. As can be seen in the Table, natural gas is an expensive alternative to electricity, as is the price of fuel oil when used for heating. However, the use of these alternatives should be seen as a risk premium that households are willing to pay for the availability and potential stability of these alternative fuel sources. The average spent on energy resources without including electricity is approximately 50 euros per month

Conversion table for energy factors can be found in Table 8, these are estimates, as the efficiency of the appliance will impact the actual kWh equivalence. 25

12/3/15


P.38

(including electricity it’s €88.25). The average expenditure on electricity is €43.30 and on wood fuels €46.74 (Table 15). Table 15. Average Energy Expenditure per Month in winter on Different Energy Sources

Average Min Max

€ Share of Total € € Share of Total

Electricity

Wood Fuel

Small Logs

Coal

Fuel Oil

Natural Gas

43.30

46.74

41.09

25.14

8.14

9.74

55%

39%

1%

3%

0%

1%

3.77 295

0 285.71

0 100

5 68.57

1.71 14.57

0.57 60

100%

100%

69%

61%

32%

42%

Energy Monthly Costs 88.25

8.54 302.85

Table 16. Prices for Energy Sources 2014/15

Firewood Small logs (25-50cm long) Lignite Fuel Oil Natural Gas/LPG

Unit

N

Mean

Std. Dev

m3

537

26.72

13.92

30.00

0.00

48.00

ekWh Mean Price 0.04

m3

11

26.27

10.30

30.00

0.00

35.00

0.04

mt Liters Liters

57 3 69

34.06 0.71 3.94

29.19 0.28 10.32

25.00 0.60 0.60

2 .50 .40

200 1.02 70.00

0.02 0.06 0.60

Media n

Min

Max

Depending on how energy consumption and income are calculated, then the amount of income spent on energy can vary. The assumption on income only includes stated income, not remittances. When looking at the monthly average expenditure for wood fuels, coal, natural gas, fuel oil for the seven months of the winter season and then dividing by 7 to get a month average, household expenditures can be broken down into category (Figure 9). As can be seen in Figure 9, the homes with expenditures on energy less than 50 euros/ month are more likely to consume only electricity or to substitute with ‘free’ wood fuels harvested from their own forests. Of those that harvested from their own forests, 126 indicated a 0 price and therefore are not captured in this figure. The trend indicates that households tend to only spend half of their expenditures on electricity.

12/3/15


P.39

350

Euros per month (winter)

300 250

Natural Gas Fuel Oil

200

Coal 150

Small Logs Firewood

100

Elecricity

50 0

Share of Monthly Energy Expenditure by Fuel Type

Figure 9. Average Energy Expenditure per Month in Winter on different Energy Sources Figure 9 highlights that household expenditure on energy sources varies considerably throughout the population. Switching between and stacking these fuels happens at almost all levels of expenditure. Theory would dictate that the choice of energy source would be dependent on the next available source. Unexpectedly, the price of wood fuels and the time before the harvest season varied (Figure 10). There was also no particular pattern with respect to village, quantity or if the wood was cut, to explain the price variation in the season. One significant issue is that the usage-costs for alternatives to electricity are fixed prior to the heating season. The stock of wood fuels and lignite is prepurchased and should be considered a sunk cost to households when making the electricity consumption decisions month by month. However, if the house runs out of fuel wood, the cost of electricity is the highest of the year, representing a significant premium on wood fuel management.

12/3/15


P.40

Figure 10. Seasonal Prices of Wood Fuels26 When comparing the consumption of the different populations of the LPLS data to the rural averages, we can see that the lack of reliability of electricity increases wood fuel consumption by approximately 1 solid cubic meter annually, and for households that have issues with payment the consumption increases by 2 solid cubic meters. Table 17. Average Annual Wood Fuel Consumption for Kosovo Sample Population Solid m3 Rural 8.2 Urban 6.7 Low-payment (LP) 10.35 Low service (LS) 9.2 27 Sources: The impact that overharvesting is having on Kosovo’s forest, of 0.8 million estimated cubic meters28, therefore moving consumers into electricity may have a positive impact on the overall impact of energy consumption on the environment . Perhaps one alternative for electricity to compete is to give households a flat consumption rate, where slightly higher The numbers 3333 indicate ‘it varies’, and 4444 indicates that ‘it depends’ Bowen et al., “Kosovo Household Energy Consumption: Facts and Figures”; UNFAO and CENR, “Wood Fuel Consumption in Kosovo Households.” 28 UNFAO and CENR, “Wood Fuel Consumption in Kosovo Households.” 26 27

12/3/15


P.41

bills in the summer subsidize winter consumption. The rates could be set based on previous years of consumption and balanced at the end of the season. 4.2.2.

Quantities of Alternative Energy Sources

4.2.2.1.

Electricity Electricity consumption was asked in terms of the amount that the household paid

during the months of July, October through March. Households mid-survey were receiving their April bills and could report the amounts if available. The kWh consumption for the household was backed out of the bills reported, less VAT, for the digital customers to give an approximate amount consumed. The share of electricity in terms of kWh for the sample can be found in Figure 11. The average share of electricity was 34% with a standard deviation of 17%. The average kWh consumed for digital customers was about 520 kWh for the winter months. As with te interpretation of the energy graphs for other energy sources, as the composite of consumption factors increases, household trade away from electricity as an energy source. Additionally the average consumption for energy in the household is about 1,450 kWh, using the average monthly consumption for winter for digital customers and the converted ekWh for other energy sources. When considering their heating most households in the sample split the consumption of energy resources between electricity and wood, which can be seen as the diagonal 45 degree line (Figure 12). This also indicates the dependency of the households on wood fuels over the electrical system.

12/3/15


P.42

Figure 11. Scatter Plot of the Share of Electricity for Digital Meter Consumer by Total Monthly Consumption (kWh)

Figure 12. Scatter Plot for Digital Meter Consumers for the Share of Electricity by the Share of Wood Fuels (kWh)

12/3/15


P.43

The household income levels were compared to the average monthly electricity bills. Income affects the average amount that households are able to spend over the winter months. In the month of January, those with digital meters consumed an additional €3.61 for every €100 of income level increased. For the analog meters, consumption increased by €0.98 for every income class increase. It can also be seen from the figures that there are some discrepancies in the analog meters in the expected monthly bills. For example, those that have no income have higher monthly bills than their digital counterparts (Figure 13 and Figure 14). This result indicates that there is an issue with the price signal that these households see (consistent with the idea that the analog meters have been tampered with), as it would be consistent that the higher incomes would spend more than lower income homes. This may make sense that the households on the analog meters may be consuming more kWh at a lower price, and making their bills higher than they would if they were getting an accurate signal. However, there may be other explanations outside of this.

Average Monthly Bills (Euro)

120.00 100.00 80.00

Digital y = 3.6057x + 35.064

60.00 40.00 20.00 -

Income Category Oct

Nov

Dec

Figure 13. Electricity Expenditure by Digital Meter Type, Income Level and Month (€)

12/3/15


P.44

80.00

Analog

Average Monthly Bill (Euro)

70.00 60.00

y = 0.9858x + 41.823

50.00 40.00 30.00 20.00 10.00 -

Income Category (Euros) Oct

Nov

Dec

Jan

Feb

Mar

Linear (Jan)

Figure 14. Electricity Expenditure by Analog Meter Type, Income Level and Month (€) 4.2.2.2.

Wood fuels

Assuming the following rates of conversion29, 1 m3 of fuel wood is approximately between 734 kWh of electricity. Estimates for the energy equivalent value of wood fuels is usually between 1,400 and 2,000 depending on tree species and drying time to get to a 20% moisture content30. The most common species of wood found in Kosovo used for fuel wood is oak and beech31. Likely given the short time between purchase and the heating season (70% of the 1160 surveyed in the FAO work purchased less than 2 months ahead) the moisture content is likely around 60%32. Multiplying the market price for wood fuels by this conversion rate indicates that the wood equivalent price is between 0.02 to 0.067 Euros per kWh. When compared to electricity tariffs for the winter months in lower block

291

cubic meter = 0.3531 metric ton; 1 metric ton fuelwood = 0.3215 tons of oil equivalent (toe); 12 MWh toe; UNECE, “Forest Product Conversion Factors: Project Overview and Status.” 30 Biomass Energy Center, “Typical Calorific Values of Fuels.” 31 UNFAO and CENR, “Wood Fuel Consumption in Kosovo Households.” 32 Green Oak is estimated to have a moisture content of 75-80%, to get to the 20% in which the kWh equivalents would be higher, would take approximately a year, with low ambient humidity. USDA, Wood Handbook: Wood as an Engineering Materials.

12/3/15


P.45

tariffs33, consumption of wood is more expensive than electricity at market prices above 39.5 Euros per m3. The prices ranged from 15 to 43 €/ m3 depending on when the wood fuel was purchased. However, this price conversion does not include wood fuels that are used for multiple activities (e.g. heating and cooking) which would increase the relative value or opportunity cost of using the wood fuels. To calculate the amount of wood that is consumed for the winter heating months, two adjustments were made. For households that purchased their wood, the reported amount was adjusted by a standard factor of 0.69 to get the solid wood amount34, as there are losses in cutting and stacking. Households that harvested their own forests the amount was not adjusted. Additionally, households that use wood year round average 2.08 m3 less than the households that only use during the heating season, to get the winter amount for these homes, the quantity was decreased by this amount. Additional factors that would impact the efficiency of the wood to kWh conversion would be the condition and age of the stove and the need for maintenance. On average, there was 9.76 m3 of solid wood fuels consumed for the entire season in 587 of the surveyed households. Additionally, many households (n: 480) surveyed only used wood fuels for heating. This equates to roughly 1,040 kWh of energy monthly in the winter for the household displaced by wood fuels for those that do not use any heating devices during the winter months. For households that substituted out some of their heating with electric devices the amount of energy displaced by wood fuels was 961 kWh monthly. When adding together all of the households in the sample the total amount of energy displaced is 599 MWh in wood fuels, during the 7 months of winter, given 2014/15 temperatures. If the wood fuels were properly dried, the amount could potentially be double.

33The

average winter tariff for electricity in Kosovo in 2015 is 0.042 for the block of 0-200 kWh per month consumed, 0.058 for the block of 200-600 kWh per month consumed and 0.084 for greater than 600 kWh per month consumed, which can be found on the back of any KEDS bill or in ERO, 2014 Tariff Rates for Kosovo.. 34 UNECE, “Forest Product Conversion Factors: Project Overview and Status.”

12/3/15


P.46

Table 18. Descriptive Statistics of Wood fuel Consumption Question

N

Mean

Median

Min

Max

392

10.46

10.00

2.76

34.50

(solid m3)

210

8.47

6.90

1.38

69.00

Low-Payment

(solid m3)

292

10.35

9.83

2.76

34.50

Low Service

(solid m3)

292

9.20

7.59

1.38

69.00

Those that don’t' use an electric device for heating

(kWh)

480

1,040.84

868.22

144.70

7235.14

Those that use an electric device for heating

(kWh)

125

961.34

867.92

289.11

3145.42

When do you use wood fuels

Monthly ekWh for wood for the winter season (adjusting off 2 m3 for those using all year)

All year round Only the Heating Season

Unit (solid m3)

Households that were in the low-payment areas of the survey are slightly more likely to also be those homes that use wood fuels year round. The average amount consumed for these households is 10.35 m3. Households that just experienced low-quality service had winter only average consumption of wood fuels at about 9.2 m3. It is likely that many of these households would not switch completely away from using wood fuels, even as the price dynamics change with the price of electricity. The UNFAO and CENR work estimated that the amount of wood fuels consumed annually in households was approximately 2.05 million cubic meters annually (an average of 8.24 m3)35; sustainable harvest limits are estimated to be about 1.3 million cubic meters annually36; for the forest sector to be sustainable, the energy system and efficiency efforts would displace the difference 0.75 million cubic meters or 62.8 MW of capacity.

35 36

UNFAO and CENR, “Wood Fuel Consumption in Kosovo Households.” Norwegian Forest Group, “Kosovo National Forest Inventory 2012.”

12/3/15


P.47

The share of wood fuels in terms of kWh for the sample can be found in Figure 15. As monthly consumption of energy increases, the share of wood fuel consumption increases. The curvature patterns in the shares indicate an increase of a composite of other consumption factors (e.g. number of people, income, etc.) leads to a greater share of wood fuels consumed with diminishing marginal returns. The average share of wood fuels was 61% with a standard deviation of 18%. Almost all of 605 households used wood fuels as an energy source, sans 8 homes that did not know the quantity or price of wood fuels purchased.

Figure 15. Scatter Plot of the Share of Wood Fuels for Digital Meter Consumers by Total Monthly Consumption (kWh) The average amount of time before the heating season that wood was harvested or purchased was 1 to 2 months (70%, n=432). Homes typically used wood for cooking (n:6), heating (n:11) or both (n: 577). The source of wood fuels was mainly from traders (n: 410) or from own forests (n:143) or a combination of both (n:29). Those that harvested from their own forests indicated often that there would be no price at which they would switch from wood, as they were getting the wood for free (often excluding the costs of harvest, and likely not replanting). Those that do collect wood fuels tend to be men (n:322; with only 6 women and 12 families reporting both) and 77% were over the age of 40.

12/3/15


P.48

4.2.2.3.

Lignite

Lignite was consumed in 57 of the surveyed homes, in order to get a kWh equivalent, the quantities of coal were converted by a factor of 1 ton lignite Coal can be converted to approximately 1.83 MWh. This is the conversion factor for coal to energy used for the lignite power plants at 27% of thermal efficiency, assuming that the energy content of the lignite found in Kosovo is 6,942 BTU/kg. The conversion factor was then estimated to be approximately 70% of what could be produced at the plants, given that household stoves are not like to be as efficient as even Kosovo A or B. The share of lignite in terms of kWh for the sample can be found in Figure 11. The average share of lignite was 47% with a standard deviation of 18%. Lignite in the homes is being used mainly for heating (n:44) or for both cooking and heating (n:34), very few (n:5) use it only for cooking. Coal is almost exclusively purchased by men, except in 2 homes in which both genders could purchase.

Figure 16. Scatter Plot of the Share of Lignite for Digital Meter Consumer by Total Monthly Consumption (kWh) 4.2.3.

Income and Expenditure on All Energy Sources

Survey respondents indicated that the average monthly kWh consumption of energy source that are not electricity is approximately 1,191 kWh monthly in the winter (Table 19). The

12/3/15


P.49

outlier in the data was one household that reported consuming 100 m3 of wood fuels, and was estimated to consume 7,235 kWh of energy resources monthly, the upper 75% quartile consumed 1,447 kWh and the lower 25% of the sample consumed 723 kWh monthly. Table 19. Descriptive Statistics of Non-Electricity kWh Consumed N

Valid Missing

Mean Median Std. Deviation Minimum Maximum

590 15 1,191.15 1,085.27 680.71 144.70 7,235.14

As the percent of household income spent on energy sources increase, the homes in the 20-25% of income spent on energy, almost double their consumption of coal relative to other sources (Figure 17). Also considering households under the 10% expenditure level consume 2,500 kWh monthly in the winter, whereas the homes that are expending more euros on energy are also consuming more energy in kWh overall. 4000.00 3500.00 Monthly kWh

3000.00 2500.00 Natural Gas Fuel Oil Lignite Solid Wood Adjusted Average Winter Electricity

2000.00 1500.00 1000.00 500.00 0.00 0 - 11 - 16 - 21 - 26 - 36 - 46 - >56 10% 15% 20% 25% 35% 45% 55% Percent of Income Spent on Energy

Figure 17. Energy Consumed by percent of income spent on energy Coupled with Figure 18 and Figure 19, the situation indicates that household income relative to energy consumption for different sources, although may be increasing, does not

12/3/15


P.50

lead to an increase in modern electricity consumption. Households that have more income tend to consume greater portions of their monthly energy consumption on lower value

Monthly kWh

fuels (e.g. lignite and wood). 4500.00 4000.00 3500.00 3000.00 2500.00 2000.00 1500.00 1000.00 500.00 0.00

Natural Gas Fuel Oil Lignite Solid Wood Adjusted Average Winter Electricity

Income Categories

Figure 18. Average Energy Consumed (kWh) by Income Category When looking at the amount that the households spend on energy fuels in the winter months, the overall expenditure is between €40 and €70, with expenditure only increasing by about €1.93 for every additional income category (Figure 19). Between the two figures we can see that although the amount spent is not increasing rapidly, households are

Monthly Expenditure (Euro)

choosing to substitute for other fuel sources. 200.00 150.00 100.00 50.00 -

Income Category Mean

Minimum

Maximum

Figure 19. Average Monthly Expenditure by Income Level (€) 12/3/15


P.51

4.2.4.

Affordability

One of the arguments that is frequently raised in the energy debate is the affordability aspect of energy with respect to the household budget. The hypothesis for these areas surveyed may be that they represent some of the poorest households within Kosovo and would qualify as energy poor – defined by the percentage of income spent on energy, access to reliable energy services and main sources of energy fuels. Additionally, hypothesizing that these households which may be just increasing their debt burdens and unable to afford a minimum level of energy. However, the standard that needs to be defined is what the threshold of affordability is in terms of both kWh and ekWh (which include other sources of energy converted to kWh equivalents), and in terms of euros spent. Additionally, when considering the substitution of energy inputs, there are significant trade-offs considered between improving energy efficiency (getting more usage per kWh) and conservation (decrease of energy consumption). Both impacts could decrease the amount of kWh needed for a household to maintain a standard of living. Energy poverty is a delicate subject to tackle, when considering the amount that a household spends on energy in relation to the household’s overall income and expenditures. As it is difficult to accurately ascertain income and savings, as in developing countries the inflows of remittances and non-traditional forms of savings (e.g. livestock and other financial tools) need to be considered in relation to household outflows. One rule of thumb to measure energy poverty is the 10% rule, where a household should not spend more than 10% of income on energy resources37. When looking at the percentage of household income that goes to energy resources (amount spent per month added together and then divided by income), there are several households that are under the 10% metric that defines energy poverty (Figure 20). The category ‘all energy’ includes the average winter electricity bill, and the monthly amount spent on wood fuels, coal, fuel oil, and natural gas, divided by monthly income reported (excluding remittances). The category winter and summer average are the winter and summer electricity bills divided by monthly income. And lastly, the self-reported category was the percentage that households felt they were spending on average per month on energy resources, when asked directly. The missing category is people that did not report their income levels or reported as a range.

Schuessler, “Energy Poverty Indicators: Conceptual Issues Part 1: The Ten-Percent Rule and Double Median/Mean Indicators.” 37

12/3/15


P.52

Number of Households

300 250 All Energy Sources winter consumption

200 150

Electiricty Average Winter Bill

100

Electricity Summer Average Bill

50 0

All Energy sources - Selfreported

Percent of Income

Figure 20. Percentage of Income Spent on Energy by Number of Households 4.2.4.1.

Assistance

According to interviews done with Nexhat Syla, an official of Ministry of Labor and Social Welfare and Agim Krasniqi, the director of the financial budget from the Ministry of Finance as part of the pre-survey work that was done by Dina Vllasaliu there are some social assistance programs available for energy consumption in households. According to Syla and Krasniqi, from 2005, around €4.5 million are being allocated to the general social assistance scheme that goes both to families with low income (€1.72 per day per adult) as well as families of martyrs, war invalids, and civil victims. The number has not been adjusted for increases in electricity prices, or inflation. The assistance fund for electricity is a part of the general social assistance scheme and covers the first 400 kWh of monthly consumption and is not adjusted for increases in electricity prices (which with the given threshold of 400 kWh would mean fewer families are being covered)38. Within the sample of households surveyed 5 households currently participate in the program, 5 households have participated in the program and 29 households were aware of the program, but do not or have not participated. All of the households enrolled were under the €300 per month income bracket and have between 3 and 9 people residing in the Dina Vllasaliu, “Direct and Indirect Factors Affecting the Regular Monthly Payment of Electricity Bills in Hajvali.” 38

12/3/15


P.53

household. Since there are not more than 5 individuals in each category, any statistics would not be representative for an overall analysis. However, based on a 30 day month, 31% (n=182) households of those surveyed are under the €1.72 per day per adult metric. These homes tend to have larger family sizes and be making under €500 monthly (Figure 21). These households could potentially be on social assistance, but perhaps do not know that they could qualify or there is not enough funds to cover the additional households. These households might also be able to decrease consumption through efficiency programs targeted at houses under the poverty line.

Figure 21. Social Assistance Metric by Income and Total People Those that could be on social assistance spend on average €7 less on electricity than those who would not qualify based on income (Figure 22). Additionally, subsidizing the first 400 kWh may not be the most efficient policy. When calculating out the kWh difference on average, to raise the households to same level as those that would not qualify is a difference of 85 to 125 kWh monthly (Table 20).

12/3/15


P.54

Monthly Bill (Euro)

60.00 50.00 40.00 30.00 20.00

Above Social Assistance Threshold

10.00

Social Assistance

-

Month

Figure 22. Households Qualifying for Social Assistance Table 20. Consumption Difference for by Social Assistance Metric (kWh, digital Meters)

480.50 518.42 568.03 637.05 584.66 566.64

Under 1.72 per person per day 370.65 392.62 452.46 551.29 468.95 471.37

559.22

451.22

Not Under 1.72 per day October November December January February March Average Winter (kWh)

Difference 109.84 125.80 115.58 85.76 115.71 95.27 107.99

When considering households that are on the social assistance metric of €1.72 per person per day, these households consume less energy in terms of electricity and lignite relative to those that are not under the metric (Figure 23). The difference between the averages of those that could be on social assistance and those that are not is a difference of less than 500 kWh, most of which comes from the consumption of lignite.

12/3/15


P.55

kWh

3500.00 Natural Gas

3000.00 2500.00

Fuel Oil

Monthly

2000.00 Lignite

1500.00 1000.00

Solid Wood Adjusted

500.00 0.00 Not under 1.72 pp/day (N=414)

Under 1.72 pp/day (N=191)

Average Winter Electricity

Figure 23. Energy Consumed by Households That Would Qualify for Social Assistance 4.2.5.

Debt

One aspect of affordability is the long term sustainability of household energy debt relative to the percentage of income that the household is expending on energy resources. Of the total households surveyed, 164 said that they had an outstanding balance with wither KEDS/KEK. There can be several factors that account for being unable to afford one’s bills. If we assume that as income spent on energy resources increases that the affordability and the likelihood of debt may increase. As seen previously, the source of income does not give a stark indication on the type of household that will have debts. For households spending under 15% of their income on all energy sources, there are less people that reported having debts. However, as the percentage of income spent increases – there is no difference in the number of households (Figure 24). To decrease the debt, one option may be looking at reducing consumption to under 15% of a household’s monthly income. This may decrease the probability of those households from being nonpaying to paying customers, because as the percentage of income spent increases over this threshold the likelihood of being in either category is equal or greater. Those households that are spending over 35% of their income, likely are part of the structural poverty issues in the country.

12/3/15


P.56


70 60 50 40 30

No

20

Yes

10 0 0 - 10% 11 15%

16 20%

21 25%

26 35%

36 45%

46 55%

>56

% of Income Spent on Energy Source

Figure 24. Income Spent on All Energy Resources and Whether or Not the Customer Has Debt Potentially when considering the income level and whether or not the household has debts or not, there is not a reasonable difference between households in the same income classes, except in households under 100 euros per month (Figure 25). Debts are also not significant to the amount of electricity that is consumed on a monthly basis, indicating that those with debts are no more likely to increase consumption than those without.


60 50 40 30 20

No

10

Yes

0

Income Categories

Figure 25. Income Levels and Debt

12/3/15


P.57

Factors that were more likely to be positively correlated (significant under 10% levels) with debt were, the total number of people in the house, the number of people under the age of 18, if their main source of income was social assistance, if their walls or roofs were insulated, if there were leaks around their windows, the size of the electricity bill for October and March, if they communicated with KEDS employees when meters were being read, if they didn’t pay their bills at the KESCO windows and those that have problems paying their water and electric bills. Given that the normally expected factors did not end up as conclusive in terms of predicting whether or not the household has debt, Pearson correlations were calculated for a variety of different variables in the system. Factors that were significant under 10% levels and negatively correlated with debt were: education of the head of household (more educated heads were less likely to have debts), the total amount of money spent on nonelectric sources, the type of meter (analog meters, were less likely to have debts) and whether or not they were more likely to say ‘they’d invest in energy efficient appliances, but had a difficult financial situation.’ This indicates that the energy budget as a whole not just electricity should be considered. The changing prices from the analog meters to the digital meters should be looked into further as an issue with affordability. For households to spend the equivalent amount from the single-tariff to the two-tariff system they would need to consume 70 percent of their total hours during the day time and 30 percent of their hours at night. Although, when considering some of these factors with energy consumption and the poverty metric, more analysis is needed on the welfare and expenditure impacts of altering these variables. For example, insulating homes result in conflicting welfare if greater space is heated it can result in greater expenditure on electricity and other energy sources consumed, otherwise called the rebound effect. However, if the same space is heated, the displacement of expenditure on energy sources may lead to a repayment of bills or consumption in other areas. Based on the data found in this survey, even as people fully isolate their homes the impact may be minimal with on average only 3% more of the surface area of the home being heated. The systematic issues related to poverty may remain, though there may be nonmonetary gains in welfare and quality of life. Additionally, focusing on households that are at the threshold of affording their bills may be the most economical in terms of household interventions. These are households that, during the heating transition months of October

12/3/15


P.58

and March, have issues paying their bills. The focus area of the study is households in low payment and low service areas, where interventions on many levels may be required. The simple fixes in their homes to decrease heat losses may make an impact on their abilty to pay their bills. 4.3. KEDS and the Consumer 4.3.1.

Bill Understanding and Electricity Consumption

A common form of uncertainty about non-price attributes emerges in the literature regarding household behaviors on consumption, where services are consumed. Electricity is an input to other aspects of consumptive behavior. For example, we watch television, but we do not think of the amount of electricity that goes into this activity. Durable goods (e.g. office equipment, appliances, and televisions) requiring energy or water inputs fall into this particular category. This lack of information leaves people uncertain as to how ordinary actions like heating or cooling a house by a couple of degrees translate into electricity usage. Generally, poor information leads to inefficiency as households fail to understand the energy requirements for appliances and usage. This includes a fuzzy understanding regarding not just the price of energy and the costs of energy efficient investments in order to equate the marginal benefit and the marginal cost39. Decisions can be altered when information is communicated in a simple manner especially as new technology is available for monitoring consumption40. When looking at the experiences of government transfer programs’ take-up rates or the amount of eligible individuals accessing the program, and how it increased when information was delivered in a simple to understand style41. This process can also be seen in other social intervention programs like retirement plans42, consumption in cell phone minutes when users are informed of a higher price tier approaching43. Developments in dashboard displays in cars have increased the knowledge of drivers of the gasoline needed to travel distances44. In order to gauge the understanding of consumers and the services that they receive from electricity several questions were asked, both directly and indirectly. One of the questions in the survey which relates to the consumer’s direct understanding of their bills Jessoe and Rapson, “Knowledge Is (less) Power.” Ibid. 41 Bhargava and Manoli, “Why Are Benefits Left on the Table?” 42 Duflo and Saez, “The Role of Information and Social Interactions in Retirement Plan Decisions.” 43 Grubb and Osborne, “Cellular Service Demand.” 44 Stillwater and Kurani, “Goal Setting, Framing, and Anchoring Responses to Ecodriving Feedback.” 39 40

12/3/15


P.59

was “Do you understand the bill that you receive from KESCO?” The interviewees could answer by “Yes,” “No,” or “Partially.” Then they were then asked to answer the following question “If not, what would make understanding the bill easier?” The purpose of the first question was to determine whether customers understand their electricity consumption through their bills. Whereas the latter question asked for a specific factor, that if changed, would make their understanding of the bill easier. Another question in the survey concerned with the customers’ electricity expenditure was “Do you know that the cost of electricity varies by time of day and season?” As this was also a closed-ended question, the interviewees had the option of answering with either “Yes” or “No.” The aim of this question was to discover whether individuals were aware there are price differences depending on the time of the day and season – the electricity is more expensive during the day and during winter, as opposed to during the night and during summer. When considering the household’s understanding of their bill coupled with the type of meter, there was a noticeable difference. Figure 26 shows the difference in households with digital meters. These households indicated that they understand their bills and tended to spend less on electricity on average, than those who are not aware of this fact, or even those partially aware. This difference in spending can be seen in the winter months of October, November, December, and February. Households which understand their bills spend less. In these months it resulted in €3.61 less in October than those that do not; €4.27 less in March; €2.19 less in December; and €5.29 less in February. 60.00

Average Bills

50.00 40.00 30.00

Digital | No Understanding Digital | Understanding

20.00 10.00

Digital | Partial Understanding

-

Figure 26. Impact of Bill Understanding on Average Monthly Bills (€) for Digital Meters

12/3/15


P.60

Households indicating they partially understand their bills seem to have higher electricity bills July through March, than those who indicating to understand them. This difference ranges from €1.07 to €10.81. Based on the survey questions, these houses indicated through a follow-up question where they have an issue with understanding. These responses were grouped into three main categories: problems understanding their consumption, tariff structure, and the bill presentation. A group of individuals wanted more personalized explanation of their bills – they want more interaction with KEDS/KESCO employees to explain their bills directly to them. When it comes to households with analog meters (Figure 27), there is no real, discernible difference in expenditure between those who understand their bills and those who do not. The people who partially understand their bills seem to have lower per-month expenditures. These households with partial understanding of their bills may have switched to other energy sources to avoid higher electricity bills.

60.00

Average Bills

50.00 40.00 30.00

Analog | No Understanding Analog | Understanding

20.00 10.00

Analog | Partial Understanding

-

Figure 27. Impact of Bill Understanding on Average Monthly Bills (€) for Analog Meters 4.3.2.

Electricity Price Variation Understanding and Electricity Consumption

Tariffs vary by season and by time of day45, and the different prices are presented to the consumers on their monthly electricity bills in addition to being available online. However, 45

ERO, “Statement of Security of Supply for Kosovo (Electricity, Natural Gas and Oil).”

12/3/15


P.61

households may not be fully aware of or understand these different prices. Households aware of the tariff variation tend to alter their electricity consumption during these low tariff periods, leading to reduced electricity bills. In comparison to households not aware of this fact (Figure 28), this difference in expenditure is particularly noticeable in the bills during winter months. During this period households may decrease electricity consumption by switching to heating and cooking with wood fuels. Households that understand their bills spend approximately 5 euros less in electricity bills than those that are not aware. When it comes to households equipped with analog meters, the difference between households which understand that the electricity cost varies and those that do not is more subtle and not statistically significant. 60.00

Average Euro per Month

50.00 40.00 30.00

Understands (digital) Doesn't Understand (Digital)

20.00

Understands (analog) Doesn't Understand (analog)

10.00 -

Month

Figure 28. Impact of Electricity Price Variation Understanding on Average Monthly Bills for Digital and Analog Meters Another question in the survey concerned with electricity tariffs was “Do you know that there will not be a tariff increase for 2015?” The interviewees could answer with “Yes” or “No.” Of the 605 households surveyed, 506 indicated they were ‘not aware’ the tariffs would remain the same, 93 households were ‘aware’, while information on 6 houses was missing. This result shows that households are not informed when it comes to energy decisions made at the government level. This information was printed on the bills,

12/3/15


P.62

however, we had no question on the survey regarding the household’s comprehension of these notices. To gauge whether the consumers understand the longer term changes in prices they were asked if they understood that the cost of electricity varies by season and day and night. The amount of bills, all months were statistically significant and negatively correlated at the 5% level, except for February (significant at the 10% level) and March and April (which were not significant). The correlations ranged between -0.36 and -0.44, indicating that households were more likely to decrease consumption to a point if they understood the price differences. However, with the low correlation between both of these measures, it means that understanding prices can only impact a change in behavior to a point. If the household knew whether the tariff changes by season/time and the understanding of their bills, the correlation was -0.35 (significant at 5.2% level). However, correlating the bills and whether or not they understood the bill, none of the months were statistically correlated with less than a 10% confidence interval, when controlling for type of meter. There is also a disconnection between knowing that the tariff rates change and understanding the bill, as the two variables are not significantly correlated. 4.3.3.

Improvements to the bill understanding

Fifty-two households provided us with ideas on how to improve their understanding of the bill. Nine people responded with concerns on their consumption; twelve had issues with the bill presentation; twenty-five needed explanations on the tariff structure; while six people wanted a more personalized explanation of their bill. Some of the information is already presented to customers, such as, 

One household suggested the bill should inform the customers on the amount and price of electricity spent during the day and during the night, separately



Should clarify the amounts of electricity spent during low and high tariff periods.



Customers wanted more information on the prices for each tariff.



Some people indicated they would like to see only the total debt in the bill.

These are areas in which perhaps testing different presentations of the information might be useful. Households concerned with the bill presentation suggested a more simplified

12/3/15


P.63

receipt with a bigger and clearer font size. Those households that asked for a more personalized explanation of their bill requested more communication and consulting with KEDS employees in regards to the amount and price of the electricity spent. Some even suggested TV advertising as a means of disseminating information on how to improve the bill understanding of the citizens. The majority of households concerned with the tariff structure presented in the bill indicated they do not understand the tariffs (especially the block tariff) associated with their consumption because there are too many and too complicated. One recommendation for this issue was to only have one tariff. Some asked for more explanations on the kW spent and their respective price. To improve bill understanding, examples from other countries may be useful. Household energy consumption monitoring in the U.K. revealed that feedback relating to environmental or financial costs provided to these homes were influential in reducing their electricity consumption46. Similarly, the results of qualitative interviews in Danish households discovered that detailed feedback on their electricity consumption displayed on small LCD screens made their consumption more salient and visible. After the 5-month testing period, the average electricity saving was estimated to be 8.1% in the participating households, as opposed to the 0.8% saving in the control group. As such, feedback stimulated the electricity consumers to lower their consumption47. Another study regarding feedback’s role in reducing energy consumption found that the savings on electricity range from 1.1% to more than 20%. Usual savings tend to be between 5 and 12%48. In a similar study in Sweden, it was found that the customers’ preferred ways to receive consumption feedback were mostly in-home displays and letters. Both money and the environmental factors were significant motivational factors for saving energy, as customers were striving to lower their electricity consumption, and maintain it low49 . The outcomes of these studies illustrate the power of information to impact household energy conservation.

Gwendolyn and Alan, “Reducing Household Energy Consumption: A Qualitative and Quantitative Field Study.” 47 Alice and John, “Feedback on Household Electricity Consumption: Learning and Social Influence Processes.” 48 Fischer, “Feedback on Household Electricity Consumption.” 49 Iana Nikolaeva Vassileva, “Increasing Energy Efficiency in Low-Income Households through Targeting Awareness and Behavioral Change.” 46

12/3/15


P.64

Nonetheless, in some instances, there were no savings found while utilizing this approach. In a study of Sexton et al.50, the purpose of feedback was load shifting. Customers were informed about the considerable difference between the peak and off-peak tariffs, their current electricity use, and the estimated cost per hour. A light signal in the electricity-use monitor alerted them when there was a switch between the peak and offpeak hours. The feedback informed the customers that electricity was suddenly cheap during the off-peak hours, which led to heavy load-shifting activities. Hence, the savings in electricity during peak periods were cancelled out by the increased off-peak consumption51. The challenge for Kosovo for shifting consumption to manage the load curve is the availability of supply and issues with active, intermittent-active and passive household activities. Issues with heating water in the night is constrained by the availability of water during the night hours. During the summer and winter months about 15 to 20 households were constrained with both water and electricity issues and some homes did not have running water directly to the household. Issues with space heating can be seen in the low levels of insulation, leakages around doors and windows and adoption rates of efficient windows. Additionally, the incentive structure for the distribution in shifting consumption are limited in a regulated market, where prices and revenues are tightly controlled. 4.3.4.

Communication

Given that communication with KEDS employees is correlated with the likelihood that a household carries debts, indicates that these households may be seeking out information on their bills and or about prices. Figuring out an effective strategy for providing information to those households may improve their understanding of their energy consumption, payment rates and other energy factors. 4.3.4.1.

Where do the consumers pay bills One of the questions in the survey was “Where do you pay your bills?” and

respondents could choose one or more of the six options as their answer. As shown in Figure 29, 92.6%, (n=560) indicated they pay their bills at KESCO teller windows. Fifteen respondents stated they pay their bills the bank; five at the post office; eight by money transfer; and six through direct debit; and four answered with “Other.” Sexton, Johnson, and Konakayama, “Consumer Response to Continuous-Display Electricity-Use Monitors in a Time-of-Use Pricing Experiment.” 51 Ibid. 50

12/3/15


P.65

600 500 KESCO teller windows

400

Bank

300

Post Office

200

Money Transfer

100

Direct Debit Other

0 KESCO teller windows

Bank

Post Office

Money Transfer

Direct Debit

Other

Figure 29. Location of Payment KESCO bill pay offices are the main contact point for electricity customers may be an important factor to consider when creating an awareness campaign for bill understanding and electricity consumption conservation. There are trade-offs with awareness campaigns and the effective uptake of the information with the personalization of the programs. For example, country-wide awareness may be cheaper, but less effective than door-to-door or village-to-village. However, having most customers visiting KESCO regularly, the target audience could be reached there less expensively in the form of brochures, posters, or meetings with an authorized person. Targeting this information to the appropriate demographic may be useful. The survey question “Who in the household tends to be in charge the following energy related activities: Paying Electricity Bills? Please circle gender and age” showed that males aged 40 to 59 cover 50% in general of the sample responsible for paying electricity bills (90% n=544, indicated that men pay the bills). Hence, employees at KESCO teller windows are most likely to encounter males of these two age groups. 4.3.4.2.

Where do they get energy information?

Another question in the survey concerned with where customers get energy-related information was “Rate the extent to which you consult the following media when it comes to energy efficiency or outages?” Respondents had to rate each of the media options presented (Internet, TV, Newspapers/Magazines, Radio, Brochures/Leaflets, Word of Mouth/Friends/Family, Other) on a scale from 1 to 5 – 1 being “Always,” 2 being “Often,” 3 “Sometimes,” 4 “Seldom,” and 5 being “Never.”

12/3/15


P.66

As shown in Figure 30, the media’s frequency of use for getting energy-related information can be prioritized in the follwoing order: TV, Word of Mouth, Internet, Newspaper, Brochures, Radio, and Other. Hence, 163 (N=605) customers always get their information through the TV; 140 people often get informed through the word of mouth; 67 individuals sometimes get informed through the Internet; 56 seldom get information through the newspaper; whereas 535 people indicated to never get energy-related information through the Radio. One of the reasons behind this ranking is the availability of media in these villages. For instance, almost all surveyed households had at least one TV, whereas only 10% (n=60) of households had at least one radio. Therefore, the availability and accessibility, among others, of the aforementioned media options influence the media choices customers make to get information on energy efficiency and outages.

Other Word of Mouth Brochures Radio Newspaper TV Internet

Always

Often

Sometimes

Seldom

Never

Figure 30. The Extent to which Media is followed for Information on Energy Efficiency or Outages Correlating where consumers get information and their bills was not significant, nor did it contribute to the likelihood of debt. However, when looking at the opinions about information with energy efficiency there are links to the source of the information and the overall understanding. Households that get their energy from information from the internet are 8% less likely to be concerned about energy efficiency but are 9% likely not to care about saving energy. Those that get their information from television are correlated positively (both at 11%) with understanding that electricity has become more expensive in the last 12 months and feel that their economic situation is preventing them from

12/3/15


P.67

purchasing efficient appliances. Word of mouth information correlates to positive energy efficiency understanding (9.6%), an understanding that consumption has increased (8.9%) and has become more expensive (16%). 4.3.4.3.

Where do they get energy information - Demographics

Further analysis has been conducted on the use of TV as the medium most frequently used to get information on energy efficiency or outages. In order for KEDS to target their customers effectively through television, demographic factors like ethnicity, age, family size, income, and the education of the household head are crucial. In terms of ethnicity, our survey sample (N=605) was rather homogenous – with 558 individuals being of Albanian ethnicity, 6 Bosnian, 2 Gorani, and 1 RAE; whereas ethnicity data from 38 households was missing. Out of 154 individuals who disclosed their ethnicity and indicated to “Always” consult the TV to get information on energy efficiency and outages, 151 were Albanian, 2 were Bosnian, and 1 was member of the RAE community. Out of 147 people who stated they “Often” consult the TV to get this information, 145 were Albanian, 1 was Bosnian, and 1 was Gorani. Out of 109 individuals who indicated to “Sometimes” get this information from the TV, 108 were Albanian and 1 was Bosnian. Hence, almost all who indicated to be getting energy efficiency and outages information through the TV are of Albanian ethnicity. 4.3.4.4.

Age of Respondents and the Extent to which they use the TV to get information on Energy Efficiency and Outages

In the survey age was asked in two places, the age of the respondent and the ages and activities of the household members. The responses of “Always” and “Often” consulting the TV to get energy-related information were the most common responses combined in every age category (Figure 31).

12/3/15


P.68


160 140 120 100 80 60 40 20 0 18-20

21-30

31-40

41-50

51-60

61-70

71-80

81-90

Age Always

Often

Sometimes

Seldom

Never

Figure 31. The Extent to Which Media is followed for Information on Energy Efficiency or Outages based on Respondents' Age Given the information sharing within the household, it is useful to look at the size of the family and the frequency of information. In terms of the family size, 40.9% (n=247) of households with 10 family members indicated that they get energy-related information from the TV, as shown in Figure 32. At least half of each of the number of households interviewed by every size category is getting their news from television.

Percent of Respondents

140 120 100 80 60 40 20 0 1

2

3

4

5

6

7

8

9

10

11

12 13+

Number of Household Members Always

Often

Sometimes

Seldom

Never

Figure 32. Households of Different Family Sizes Consulting the TV for Information on Energy Efficiency and Outages 12/3/15


P.69

Respondents with monthly income under €300 get information on energy efficiency and outages mostly through the TV and word of mouth. As shown in Figure 33, out of households declaring they “Always” follow one medium, 58.4% (n=353) indicated to “Always” get energy-related information through the TV. The word of mouth is ranked second, with 21.1% (n=127) of respondents indicating they “Often” get this information through the word of mouth by their family and friends. The internet and newspapers seem to be less followed, with 10% (n=60.5) of respondents “Sometimes” getting their information through the internet, and 4% “Sometimes” getting informed through newspapers. The radio and brochures are the least followed, with five respondents (1%) “Often” getting energy-related information through brochures, and two respondents (0.7%) of respondents getting this information through the radio. 300 250 200 150 100 50 0 Internet

Newspaper Always

TV Often

Brochures Sometimes

Radio Seldom

Word of Mouth

Other

Never

Figure 33. The Extent to which Media is followed for Information on Energy Efficiency or Outages by Respondents with Monthly Income under 300 euros Income may correlate to different sources of media consumed, households most frequently answering with “Always,” “Often,” or “Sometimes” to the extent they consult the TV were those with monthly income from 100 to 200 euros, households with income from 200 to 300 euros, and those from 300 to 400 euros, as shown in Figure 34. These three groups of households constitute 52.16% (n=315) of all households indicating they “Always” get information through the TV. Groups of households with higher monthly income (500 – 600; 800 – 900; and 900 – 1000, euros) were more likely to answer with 12/3/15


P.70

“Always,” “Often,” or “Sometimes” to this question. Nevertheless, the number of households belonging to these high income groups is much smaller than that of lower income groups. 100%

Percent of Respondents

90% 80% 70% 60% 50%

Never

40%

Seldom

30%

Sometimes

20%

Often

10%

Always

0%

Income

Figure 34. Households of Different Income Categories Consulting the TV to get information on Energy Efficiency and Outages Education of the household head plays a role on the household’s media choice. As seen in Figure 35. High school was the most frequent level of schooling in the sampled households. Within this group, almost three-quarters of these households often or always get their information from TV. The reason behind these results might be because household heads who completed elementary school might rely more on the word of mouth, their family and friends, whereas those with an undergraduate degree might focus on the internet or newspapers to get information on energy-related matters.

12/3/15


P.71

250


200 Never 150

Seldom Sometimes

100

Often Always

50

0 None

Elementary

High School

Undergraduate

Figure 35. Education of Household Head and the Extent to which they consult the TV for Information on Energy Efficiency and Outages Reducing Energy Consumption through Community Knowledge Networks (RECCKN) research project which investigated how people learn about energy efficiency in the U.K., found that people mostly learned from direct experience, meaning by using energy at home, or from interactions and discussions with family, friends, and neighbors52. This situation is applicable to Kosovo rural households, as “Word of Mouth/Family/Friends” resulted to be the second most popular way people get information on energy efficiency or outages, after the TV. Since interaction and dialogue are effective ways of sharing energyrelated information, enabling customers to meet with authorized KEDS personnel through regular local events could result in a success. Nonetheless, having most customers visit KESCO regularly, the target audience could also be reached there less expensively. Focused and personalized feedback, tailored to individuals’ existing level of knowledge on energy consumption and saving, would appeal more to the customers. In terms of TV, “top-down” information campaigns from large companies and institutions help raise general awareness, but because information is standardized, they are often seen as overly general and irrelevant to the everyday reality of energy use53. Further, if people do not trust the information source, then advice is less likely to be acted upon. There is a strong distrust towards advice and information from businesses and 52 53

Dobson, “Introducing RECCKN – Summary of Key Research Findings.” Ibid.

12/3/15


P.72

energy companies, resulting from media reporting and negative personal experiences54. Another downfall of using a “one-way” communication process, like the television, is that it does not provide the recipients with the opportunity ask questions and explain their situation. As such, awareness campaigns through TV could be more successful if they were to be shared by an impartial company, NGO, or expert, and also paired with another means of communication, like direct encounters with KEDS personnel. 4.3.4.5.

Websites

Another question in the survey regarding the issue of communication between customers and the energy actors was “Have you ever visited the following websites for information about outages, bills, energy information?” Respondents could answer by mentioning one or more of the following energy actors “KEDS/KESCO,” “KEK,” “ERO,” and/or “None.” However, the results from this question were particularly skewed. Only, 48 households indicated to have visited KEDS/KESCO’s website; 2 have visited KEK’s website; 1 has visited ERO’s website; whereas 91.5% (n=551) of households have not visited either one of the aforementioned websites. One of the primary reasons why this mode of communication with the customers is so underused may be the limited access to a computer – 50% (n=302) of the surveyed households owned at least one desktop, while 29% (n=175) indicated to own at least one laptop. These results indicate that websites are the least effective method of communication for various reasons, including access to a computer, access to internet, age of household head, and technology savviest of household members. 4.3.4.6.

Mobile Devices Although computers have relatively low adoption rates in Kosovo, cell phones are

highly accessible. Most households have at least 1 to 3 phones available (Figure 36). Within these phones the brands that are most common are iPhones, Samsung and Nokia, which means that there are several households that have the potential to access the mobile web or use Wi-Fi. Cell phones could be used to communicate use, outages, bill pay reminders etc. and present an additional and direct method of communicating to the customer.

54

Ibid.

12/3/15


P.73


140 120 100 80 60 40 20 0

Number of phones

Figure 36. Quantity of Mobile Phones by household 4.3.4.7.

Do they communicate with the employees

One of the questions in the survey concerned with the communication between KESCO employees and the customers was “Do you communicate with KESCO employees when they come to register electricity consumption on electric meters? (Do they tell you how much kWh you have consumed?)” Respondents could answer with “Yes” or “No.” As shown in Figure 37, 55,5% (n=336) of respondents indicated they do communicate with the KESCO employees, whereas 43.8% (n=265) respondents indicated they do not. Data from four households were missing, among which one household stated they did not know the answer to the question in hand.

12/3/15


P.74

0

50

100

150

200

Do Not Communicate

250

300

350

400

Do Communicate

Figure 37. Number of People who communicate with KESCO Employees when they come to Register Electricity Consumption Customers do not communicate with KESCO employees might impact the understanding of their electricity consumption and its respective price. Comparing the relationship between customer-KESCO communication and the customers’ understanding of the bill through cross tabulation, it was discovered that 75.45% (n=456) of the people who communicate with KEDS employees understand the bill they receive from KESCO (Table 21). The relationship between these two variables might signify that communication with authorized personnel leads to the improvement of bill understanding of individuals. Hence, having more KEDS employees communicate with customers when registering electricity consumption might lead to a better understanding of the bill, and potentially increase customer satisfaction. However, as the household bill understanding and the impact on monthly bills is not negatively correlated, it may be that the information that the households are receiving is not clear and concise, in a way that would lead to a decrease in energy consumption. One important caveat is not just that the consumers are communicating with employees but also the kinds of information that they are receiving from them. As discussed in the section on Debt, those that are communicating with the KEDS employees are also correlated positively with having debts. This could be that they are seeking out information on the debts or on how the meters are being read, the key would be to make sure that the information from the point of contact is correct.

12/3/15


P.75

Table 21. Communication and Bill Understanding Variables

H10. Do you communicate with KEDS employees when they come to register electricity consumption on electric meters? Total

4.3.5.

No

H16. Do you understand the bill that you receive from KESCO? No Yes Partially

Total

38

164

62

264

42

252

40

334

80

416

102

598

Yes

Brand Outlook

The general population of Kosovo may have a history and expectations with their energy providers. There have been several instances in the popular press that there are issues communicating between the consumer and the distribution company. It is likely that the breaking up of the energy utility from KEK into transmission (KOSTT), distribution (KEDS) and recently (KESCO) may have created some issues with how customers perceive the company. The relationship between customer-KEDS communication and the customers’ rating of the electricity distribution services before and after the privatization period was evaluated. As shown in Table 22, there was a difference between those who had communicated with employees both before and after privatization. Prior to privatization most consumers ranked the distribution company as ‘Average’ (n:245), and slightly more people communicated with the company had a positive outlook (n:117). The combination of people who had an average outlook and had communicated with KEDS employees was 23% (n=139) of the sample. Post-privatization the number of households that ranked KEDS as average was 27% (n=165), and only 83 households that communicated with KEDS had a positive outlook on the company (56 had a positive outlook and had not communicated).

12/3/15


P.76

Table 22. Relationship between KEDS Communication and Customers' rating of their Services before and after Privatization (N:590) Do you communicate with KEDS employees when they come to register electricity consumption on electric meters? No (n:256) Yes (n:333) Row N Colum Table Row N Colum Table Count Count % nN% N% % nN% N% Very Good 22 42.3 8.6 3.7 30 57.7 9.0 5.1 Good 77 47.0 30.1 13.1 87 53.0 26.1 14.8 Before Average 106 43.3 41.4 18.0 139 56.7 41.7 23.6 Privatization Bad 39 37.5 15.2 6.6 65 62.5 19.5 11.0 Very Bad 12 50.0 4.7 2.0 12 50.0 3.6 2.0 Very Good Good After Average Privatization Bad Very Bad

6 50 74 83 43

66.7 38.5 44.8 43.5 45.3

2.3 19.5 28.9 32.4 16.8

1.0 8.5 12.5 14.1 7.3

3 80 91 108 52

33.3 61.5 55.2 56.5 54.7

.9 24.0 27.2 32.3 15.6

.5 13.6 15.4 18.3 8.8

Getting feedback on their electricity consumption and expenditure on a personal basis seems to raise customers’ satisfaction of KEDS’s distribution services. The percentage of people who communicated with KEDS employees and rated KEDS’s services as “very good,” “good,” or “average” dropped after the privatization. This situation might imply that the KEDS employees are not doing as good of a job in communicating with the customers; for example, explaining the amount of electricity spent and its respective price, among others. If considering the individual changes on the household level with the outlook on the distribution system, 43% (n= 260) decreased their outlook, with most households decreasing their opinion by 2 points; 35% (n=211) had no change in opinion and 21% (n=127) had an improved opinion (Figure 38). Of those that responded 40 households went from a negative to a positive outlook, 65 remained in a negative outlook and 37 stayed within a positive outlook. However, 152 households went from a positive to a negative outlook post-privatization (Table 23).

12/3/15


P.77


250 200 150 100 50 0 -4.00 -3.00 -2.00 -1.00

4.00

Figure 38. Change in household opinion before and after privatization Table 23. Number of households and their opinions on the distribution pre- and postprivatization After Privatization Very Good Very Good Good Before Average Privatization Bad Very Bad

Good

Average Bad

Very Bad

0

1

4

13

34

1 4 2

35 57 31

23 116 19

85 49 44

20 20 8

2

5

4

0

13

4.4. Household Construction 4.4.1.

Age of Household and Surface Area

The housing stock in the sample ranges from relatively new construction to pre-conflict and older (Figure 39). The majority of homes built are from the period of 2000-2009. Older houses also have a tendency to consume more in non-electric sources of energy (Figure 40). Older houses are also slightly smaller on average than those more recently built (Figure 41). Of the LSLP data there are 4 houses that are apartments and only 10 are rented. Additionally, 21 of the houses were connected on at least one side and 7 above and below. 12/3/15


P.78

Households in Kosovo also tend not to heat their entire area, homes heat on average about 41% of their square meters, though as can be seen from Figure 42, the distribution for much lower. There are very few households (n:50) that heat the entire area. Most homes are 1 or 2 floors (n:144 and n:406) and have an average area of 116 m2. The volume of the house is on average 617 m3, though the space that is heated ranges from 15 m3 to 2,025 m3 (a 250m2 house that can afford to heat the entire space), on average 241 m3. Several factors go into the heating aspects of the energy footprint for the home.

12/3/15


P.79

Figure 39. Frequency by Age of House

Figure 41. Surface Area by Age of House

12/3/15

Figure 40. Energy Consumption by Age of House

Figure 42. Surface Area and Space Heated


P.80

Looking primarily at the materials for construction, most homes in Kosovo are constructed from the hollow clay red bricks (n:470) with ceramic tiles for the roofing. Wall width for the hollow clay bricks were 25cm, though increased to a maximum of 60cm with insulation. Solid clay bricks averaged a width of 40cm, cement brick was 25 – 35 cm, depending on insulation. The flooring in most homes is laminate, with some measure of insulation (437 indicated insulation on the floors, though likely after questioning this is merely wood boards).

Walls

0%

10%

20%

30%

Brick with Empty space

40%

50%

Solid Clay

Hollow Clay

60%

70%

Cement Brick

80%

90%

Mudbrick

100%

Other

Figure 43. Construction Material for the walls

Roof Material

90%

91%

Ceramic

92% Aluminum

93%

94%

Shingles

95%

96%

Asbestos tiles

97% Other

98%

99%

100%

Doesn't know

Figure 44. Construction Material for the Roof

12/3/15


P.81

Flooring

0%

10%

20%

30%

40%

50%

60%

70%

80%

Laminate

Parquet

Ceramic Tiles

Linoleum

Carpets

Wall to wall carpet

No Insulation

Other

90%

100%

Figure 45. Construction Material for the Flooring Homes were also asked if they had a cellar (n:180) or an attic (n:529). Those that had attics, most said that they were empty (n:351), for storage (n:147) or for living space (n:18). Only 120 of the homes indicated that the attics were insulated. Those that are empty and not insulated will spend more on maintaining heat for the home. On a positive note, 448 of the households had double pane windows, (141 had single pane, and 11 had a combination). The frames were mostly plastic (n:343) or wood (n:232), 4 homes had aluminum frames or a combination of plastic and wood (n:23). Balcony doors were also likely to be double paned (n:326) or single paned (n:109). Although the U-values are much higher with double paned windows and doors, an astonishing 269 households said that there is air that comes in around the window and/or doors of their homes. Single pane windows and double pane windows made of wood seem to have the most issues with leakage (Figure 46).

12/3/15


P.82

350 300 250 200 150 100 50 0

Single Pane

Plastic and Aluminum

Plastic and Wood

Aluminum

Wood

Plastic

Plastic and Aluminum

Plastic and Wood

Aluminum

Wood

Plastic

Air that comes in around your windows and/or doors? yes

Air that comes in around your windows and/or doors? No

Double Pane

Figure 46. Draft and Window Materials 4.4.2.

Isolation

Isolation of the household is not common in Kosovo households, as of yet, there have been some strides made on certain factors that would improve the isolation of the home to the elements. U-values for Kosovo materials have not yet been calculated; additionally issues with leakages would further compound the issue. However, it is worth looking into the kWh for homes per month spent. Although with houses built in the last 15 years, the number of homes that have Styrofoam insulation is greater than other years (Figure 47). Of the sample, 106 homes had both roof and wall insulation; 80 had walls only and 44 had roof only insulation. The homes that had their walls insulated and knew the dimensions, typically had insulation of 5cm (n:32), 8cm (n:22) or 10 cm (n:14); 14 residents indicated thicker insulation than 10cm . 200 150 100 50 0

>2010 2009 - 2000

Walls

Other

No Insulation

Glass Wool

Styrofoam

Other

No Insulation

Glass Wool

Styrofoam

1999 - 1990 1989 - 1980 1979 - 1970 < 1970 Don’t Know

Roof

Figure 47. Insulation by age of building 12/3/15


P.83

Table 24. Combination of insulation, windows and drafts Insulation Wall and Roof Walls Only Roof Only Neither Walls Nor Roof

Frequency 106 80 44 367

Although an estimate for the total amount of energy needed to heat the homes and its share of the total energy consumed needs to be calculated, the total consumption should decrease with insulation, holding other variables constant. When looking at the benefits of the insulation on the monthly kWh consumed in the winter months for all sources of energy, homes with only roof insulation fared slightly better than those with only wall insulation. The benefits of insulation for both the roof and the walls are determined mostly by the surface area of the home. As the homes increase in size, the kWh used decreases by almost 500 kWh for homes over 200m2. 3,000

Monthly kWh

2,500 2,000 1,500 1,000 500

Wall and Roof

Roof Only Electricity Bill

Walls Only

250 +

201 - 250

151 - 200

101 - 150

51- 100

0 -50

250 +

201 - 250

151 - 200

101 - 150

51- 100

0 -50

250 +

201 - 250

151 - 200

101 - 150

51- 100

0 -50

250 +

201 - 250

151 - 200

101 - 150

51- 100

0 -50

0

Neither Walls Nor Roof

Other Sources

Figure 48. Insulation by size of building and energy expenditure

12/3/15


P.84

Figure 49. Consumption of kWh by Insulation and space heated (m2) of building Figure 49 shows the consumption in kWh by space heated for electricity (blue) and other sources (green). Houses with wall insulation tend to heat more space. Houses with roof and wall insulation appear to be more efficient in their space heating and also in the kWh consumed, but a similar pattern appears when looking at homes without insulation. Additional heat and cooling can be impacted by the covering on the windows, most households had at least curtains (n:383) or blinds (n:69) or a combination of both (m:140). 4.5. Heating Systems Another impact that a household can have on the amount of electricity that it consumes is through the substitution of other energy fuels, especially in terms of heating and cooking. Households in Kosovo are known for using wood, with the Census indicating 88% of rural households as a primary source for heating55. It is not uncommon for developing countries,

55

KSA, “Kosovo Population and Housing Census 2011: Final Results.”

12/3/15


P.85

like Kosovo, to utilize fuel wood instead of electricity for heating and cooking in the household56. Urban and rural location is another significant determinant of the level of fuel wood consumption for energy purposes in households. This difference can stem from the differences in the price and availability of other energy sources like electricity, infrastructure, historic traditions, and intrinsic preferences on household heating. It was further found that greater urbanization has a negative effect on the use of wood energy57. As households move up on their status or income level, they will start utilizing a different energy mix in order to perform their energy needs58. Results from energy surveys in several developing countries claim that the majority of families have an ideal “ladder” of fuel preference running from wood fuels through kerosene to LPG, natural gas and electricity. For many rural households in developing countries there is no alternative to burning wood fuels or coal, as more modern resources are not accessible or affordable to them59. Similarly, wood-burning appliances to heat the surface of a home are widespread in Kosovo households. Wood cook stove was the most frequently used heating appliance among the surveyed households, with 577 houses owning at least one. The wood stove was ranked second, with 286 households owning at least one. Whereas, the electric quartz heater which was the most frequently used electric appliance for home heating was ranked third, with 146 households using it to heat their homes.

Song et al., “Factors Affecting Wood Energy Consumption by U.S. Households.” Song et al., “Factors Affecting Wood Energy Consumption by U.S. Households.” 58 Hosier, “Energy Ladder in Developing Nations.” 59 Ibid. 56 57

12/3/15


P.86

600 500 400 300 1st

200

2nd

100

3rd

0

Figure 50. Heating Systems Ranked by Usage In order to better gauge if households would switch to other sources of heating, the survey asked, “From the ranking above, why do you use the different energy systems? (Check all that apply).” The respondents could have multiple response answers using the seven reasons for their top three heating systems (Figure 51). The primary reason for households using either the wood cook stove or the wood stove was the cost of the energy source. Wood is generally cheaper than electricity and can be used for multiple activities simultaneously; thus, it is more affordable. The secondary reason behind the use of either a wood cook stove or a wood stove to heat their home was the quality of the heating and then the speed of the heating. For the electric quartz heater, the ranks were 1) the ease and convenience, 2) the speed of heating, and the 3) appliance’s mobility. Selection of the heating source may also have to do with the gender associated with taking care of maintaining the heat, or cooking, as activities in the household are still traditionally gender separated. The survey question “Who in the household tends to be in charge the following energy related activities: Heating/Temperature? Please circle gender and age” showed that women aged 30 to 49 are mostly in charge of the heating in the household. This means that the women are responsible for managing the household’s fuel sources for the winter months, likely also as part of the cooking and house maintenance responsibilities.

12/3/15


P.87

180 160 140 120 100 80 60 40 20 0

Electric Space Heater Portable electric heater Electric quatrz heater Gas Heater Wood Cook Stove Wood Stove Central Heating

Figure 51. Reasons behind Why Respondents Use their Top Three Heating Systems

4.5.1.

Choice of Heating and Time of use

The category of electric heating devices includes all electric heating devices covered in the survey except central heating, such as electric quartz heaters, electric space heaters, portable electric heaters, and air conditioners. Thirty-one and a half percent of the survey sample (N=605) claimed to own one or more of these electric heaters. From the sample, 76.4% (n=462) of the people who answered the question own one electric device; 15.7% (n=94) own two; 5.2% (n=31) own three; whereas 1% owns 4 and 5, the remaining 414 homes indicated no ownership. There is an outlier to these results, as one household claimed to own 9 electric heating devices. Even though a considerable number of households own at least one electric heating device, it has been found that these households do not necessarily use them. Only 20.7% (n=125) of houses use at least one electric device to heat their space, while 79.3% (n=479) do not use any. Based on the comments received during the surveying process, individuals claimed to not use electric heaters to heat their home because they cannot afford the high price of electricity, especially during the winter months. Those who stated they use electric heating devices used them in addition to wood fuel-based heating systems only for a couple

12/3/15


P.88

of hours, typically in bedrooms during the winter nights. The typical household set-up has wood heating systems located in the living rooms and kitchens. Distinguishing between day and night hours during which electric devices are used since the cost of electricity varies by time of day. As shown in Figure 52, out of the people who gave an answer to the question regarding the number of hours of daytime they use an electric heater, 66% (n=399) said they use an electric heating device for “0 hours” during this time of the day, which was the most frequent answer to this question. Households use electric heating devices more during the night time, which decreases their overall economic burden. The electric quartz heater was found to be the most used electric heating device in the surveyed households. Of the 128 households that own only one electric quartz heater, use it for 2 hours during the day and 2 during the night, on average. The 3 households indicated that they own three quartz heaters, tend to use it the most – 7 hours during the day and 6 during the night. Many respondents stated they use the quartz heater in addition to a wood-fuelled heating system. The quartz heater is used more during the night time, on average, as opposed to the day time. 120

Frequency

100 80 60 40 20 0

Hours an Electric Heating Device is Used Day

Night

Figure 52. Hours an Electric Heating Device is used The electric space heater was found to be the second most used electric heating device, 33 households indicated they own one electric space heater typically use it on average for 1 hour during the day and 3 hours at night. There were 3 households that own

12/3/15


P.89

two of these devices. It is used considerably more during the night time, as opposed to the day time, mostly because of its low maintenance and cleaning requirements. The portable electric heater is the third most used electric heating device among the surveyed households though only twenty-seven houses own one, and five own two. The typical use is for one hour during the day time. However, usage increases during the night time with two to eight hours depending on the number owned. Air conditioners seem to be the least popular choice for heating or cooling among the surveyed households. One percent of the entire survey sample (10 cases) has an air conditioner at home, which they use for 5 hours during the day and 2 hours during the night, on average. The reason behind these results may be that air conditioners are mostly used for cooling, and that is typically needed during the day in the summer months. 4.5.1.1.

Age of Electric Heating Devices The majority (56%) of all electric heating devices owned by the surveyed

households are 0 to 5 years old. Electric space heaters tend to be slightly older, 7 years old on average, with a range from 0 to 20 years. There were no portable electric devices older than 15 years. A portable electric heater is on average 5 years old among the surveyed households, with a range from 1 to 15 years. An electric quartz heater in the surveyed households is on average 6.6 years old, with a range from 0 to 24 years, with only one device belonging to the latter category. In regards to air conditioners, seven of the nine air conditioners were 0 to 5 years old, one was 6 to 10, and one 11 to 15 years old. There were no air conditioners older than 15 years.

12/3/15


P.90

80 70

Frequency

60 50 40 30 20 10 0 0 - 5 Years Electric Space Heater

6 - 10 Years

11 - 15 Years

Portable Electric Heater

16 - 20 Years

Electric Quartz Heater

21 - 25 Years Air Conditioner

Figure 53. Ages of Electric Heating Devices 4.5.2.

Average Electricity Consumption for heating devices

In order to find how much electricity on average a household consumes by using electric heating devices the following was calculated for the heating devices, (1) Table 25 shows the average electricity used for these heating devices, based on the usage and wattage capacity of these devices. From the devices that are part of the category of electric heating devices, electric quartz heater on average consumes the largest amount of electricity per month. As stated previously, the electricity consumed will likely occur at night, but the cost of operating these appliances can change depending on the other factors of household consumption and the tariff block the home has consumed up to. Table 25. Electric Heating Devices Average Electricity Consumed Electric Heating Device Electric Space Heater Portable Electric Heater Electric Quartz Heater

12/3/15

Average Electricity Consumed 78 kWh 31 kWh 116 kWh


P.91

4.6. Appliances and Energy Consumption When looking at the other households that contribute to the monthly electricity bills for the home the average amounts of kWh that appliances consume can be found in Table 26. Water heating is the appliance that consumes a significant amount of the monthly share of electricity. These averages are disaggregated throughout the rest of the section. Table 26. Average kWh consumption for appliances Appliance Water Heater Vacuum Cleaner Iron Electric Ovens Electric Hob Microwave Table-top cooker TV LCD/Plasma Digital Receiver Radio

Average kWh consumed per month 212.5 27 10.4 21.5 14.4 2 70.97 41 44 22 1.1

Households were asked the gender and age of who was responsible for purchasing new appliances. In 43% (n=260) of the households the men are the decision makers, and in 53% (n=320) of households both genders tend to decide. The main decision maker, regardless of gender, are likely to be between 40-49 years of age (31% of the sample, n=187) and then between 30-39 years (27%, n=163). 4.6.1.

Water and Electricity

4.6.1.1.

Water Boiler

Water heating is an important aspect of energy consumption in households, used for everyday activities such as showering, laundering, dish washing. Water boilers use electricity as gas is unavailable for households in Kosovo. Depending on the model, the factory set temperature ranges from 65 to 75°C

60.

Unlike other electricity consuming

appliances with fixed demands (e.g. refrigerators) water boilers and their consumption of

Gorenje, “Medium Volume Water Heaters”; Ariston, “Water Heaters General Catalogue”; Leov, “Bojler Standard.” 60

12/3/15


P.92

electricity is dependent on usage and hours turned on. When considering water heating, two areas likely to require the most consumption are showers and laundry. Both activities’ demands are based on the frequency and number of people living in the household. Therefore, in the survey, other factors of water consumption were not measured (e.g. dish washers via machine or hand). The size of household is a significant aspect of water heating for Kosovo. Water boilers in Kosovo like much of Europe, are also controlled in most homes with a wall switch, therefore most households can turn on and off the appliance when not in or preparing for use. In developed countries like Canada, water heating appliances are ranked in the second place after space heating for energy consumption61. Within the United States, water heating is responsible for 15 percent of electric usage62 and in Australia 40 percent of energy is consumed for the aforementioned purpose, with electricity being the main source (79 percent)63. In European countries, such as Norway, 14-24 percent of electricity is used for water heating purposes64. For the region, using Albania and Serbia as examples, water heating is ranked the second in terms of residential electricity consumption65. The average family size in the two countries contribute to the difference as well, with Albanian households use 20 percent of their electricity for water heating and in Serbia, 11 percent of the electric bill was attributed to heating water66. Boilers are found in almost every household - out of 605 surveyed households, 570 families or 94.2 percent owned at least one water heater. Those that did not have a boiler, either chose not to answer the question, or as some respondents indicated, their boiler was no longer functioning. In several villages, some of the respondents indicated that they were heating the water on their wood stoves for laundry and other activities as electricity was too expensive. From those that had boilers, 555 households were able to identify the age of their boilers (Table 27). The age of the boiler is going to affect the performance and efficiency of the boiler to heat water, especially given the unlikelihood that the boilers have had maintenance. Warranties for water boilers are typically 5 to 7 years and efficiency factors

Aguilar, White, and Ryan, “Domestic Water Heating and Water Heater Energy, Consumption in Canada.” Ibid. 63 Ibid. 64 Ibid. 65 IEA, “Energy in the Western Balkans.” 66 Ibid. 61 62

12/3/15


P.93

have improved considerably over the past 20 years67. Nevertheless, 60% of the boilers were older than 5 years. Table 27. Age of Water Boilers (N: 555) Age (years) 20

Number of Boilers 212 169 131 30 13

In addition to age, three factors contribute to the amount of electricity consumed, size (litres), wattage and time on (Table 28). To understand boiler usage, households were asked about the amount of time that they use their water heater during daytime and nighttime (Table 29). Household behaviors in heating water at night when prices are lower may help with the overall expenditure, however, there are limitations when access to water is not available during these times. Most houses indicated that they use their boilers 12 hours during daytime, on average 8.93 hours per day. Of those surveyed, 366 households left their water heaters on 12 hours during night-time, the average usage per night amounts to 9.57 hours, 65.4 percent left the boiler on non-stop. Although, water boilers could be on all the time, they only consume electricity when heating the tank or variably when maintaining tank temperature. Furthermore, 22 homes used their boiler 2 hours or less per day. Table 28. Water Heaters kW and Liter Range Average 2.1 82.23

Wattage (kW) Size (liter)

Min 1.1 12

Max 3.2 200

Table 29. Hours Used for Water Boilers

Day usage Night usage 67

0-6 142 107

Hours 6 -12 Average 389 8.93 435 9.57

Ariston, “Water Heaters General Catalogue”; “OGB 30-200 - Heating Systems.”

12/3/15


P.94

Of the households surveyed, 269 households could give their water capacity of their boilers and 215 households were able to identify the kW requirements of their boilers. The wattage capacity of the water heaters varies from 3.2 kW to 1.1 kW (Table 28). Some households (112 households) had a water heater with a capacity of 2.0 kW and did not specify the liter capacity. The distribution of boiler wattage can be found in Figure 54. The most common boiler size was 80 liters. Small boilers are available for kitchen usage, though only two households have a 12 liter boiler or a rating of 1.2 kW or less. Meaning that, these households do not have to leave their water heater on for longer than 2 hours to heat a full tank per day. Most boilers regardless of size are 2kW, therefore the larger the boiler the longer the time to a fully heated tank. Moreover, the largest boilers owned by 4 households have a capacity of 200 liters.


120 100 80 60 40 20 0 1.1

1.2

1.3

1.5

1.6

1.8

2

2.2

2.4

2.5

2.8

3

3.2

kW

Figure 54. Size of Water Boilers in kW

12/3/15


P.95


250 200 150 100 50 0 12

40

50

60

65

70

80

100 120 150 180 200

Liters

Figure 55. Size of Water Boilers in liters Additionally, the brand of the boiler makes a difference in factors relating to electricity consumption such as time to heat and the maximum temperature that it can heat the water (Table 30). When asked the brand of the water heater, 235 households did not know the brand, however, the ranking for the top five known brands were, Gorenje (n: 149), Beko (n: 47), Leov (n:36) and Ariston (n:15). In order to make some assumptions about the electricity consumption of the boilers the specifications for the different brands were found online (Table 30). Table 30. Specifications of water boilers

Brand Gorenje Leov Ariston

Wattage (kW)

Size (liter)

2.0 2.0 1.5

80 80 80

Time to heat (minutes) 185 110 184

Maximum Temperature (°C) 75 65 75

Based on all of this information, in order to calculate the average electricity consumption for showers in a family within a month the following calculation was used,

(2) 12/3/15


P.96

Since it was unlikely that a household would know the amount of water they consume per shower the European average per shower was used, 42.85 liters of water per person per shower (showers ranged from 26-60 liters/shower)68. It was assumed that the ratio of hot to cold water was 70:30 per shower. Moreover, using the top 5 brands in our survey, some assumptions were made about the amount of time that would be needed to heat the water, depending on boiler size (Table 30). The heating time for boilers was based on the specifications of Gorenje, Ariston and Leov water heaters69. For those not within the top 5 brands, an average based on liters and time was calculated. A water heater on average heats a liter of water per 2.13 minutes. Age was not discounted, although there might be some assumptions that the older boilers require more time to heat the water. In order to generate how many minutes on average it takes a boiler to heat a liter of water only the households that could identify the liter capacity of their water heaters were taken into consideration. As a result, by using equation (1), based on LPLS data it can be derived that a family for shower purposes on average consumes 212.5 kWh per month. Though as it can be seen in Table 31, the variation in kWh needed per month based on the number of showers per week and per person, varies. The average number of showers is 4 per person per week however, it is important to emphasize that in 108 households (19.7 percent) individuals shower 7 times a week. The direct relationship between the number of individuals, showers taken per person per week and the average electricity consumption is shown in Table 31. Meaning that, as the number of individuals within the households or the number of showers per person per week increases the average electricity consumption increases as well. For example, a household with 5 persons who each take 2 showers per week will spend on average between 0-200 kWh per month whereas, a household with the same number of individuals who each shower 7 times a week will spend on average between 200-400 kWh per month. Based on the calculations with the reported boilers and equation, there are several households that will consume an entire tariff block’s worth of electricity for heating water alone. Households may limit the number of showers based on the affordability of heating water, or take shorter than the average European shower in order to decrease these costs.

European Commission (DG ENV), “Study on Water Performance of Buildings.” Leov, “Bojler Standard”; Ariston, “Water Heaters General Catalogue”; Gorenje, “Medium Volume Water Heaters.” 68 69

12/3/15


P.97

Table 31. Average amount of kWh used, based on the number of people and the number of showers each week

Total People

1

1.5

2

1 2 34.07 3 51.11 4 65.31 5 63.89 85.19 6 51.11 98.97 7 59.63 119.26 8 136.30 9 76.67 153.33 10 170.37 11 140.56 12 204.45 13 14 15 16 17 18 19 20 25

How many showers per week per person are taken in the household 2.5 3 3.5 4 4.5 5 5.5 6 42.59 63.89 85.19 106.48 127.78 149.08 170.37 191.67 212.96

51.11 76.67 99.38 129.46 159.72 178.89 204.45 230.00 210.83

59.63 89.45 119.26 149.08 268.34 208.71 238.52

68.15 102.22 131.43 170.37 198.77 238.52 272.59

85.19

511.11

638.89

6.5

115.00 153.33 221.48 191.67 212.96 234.26 255.56 276.85 230.00 298.15 306.67 340.74 383.34 340.74 468.52 374.82

255.56 306.67

7 119.26 178.89 238.52 298.15 374.77 417.41 477.04 536.67 596.30 655.93 715.56 852.71 834.82 954.08

579.26 613.34

1073.34

85.19

12/3/15


P.98

One aspect of households in Kosovo is the fact that not all households can meet their daily energy needs for water consumption. By using the average time that a water heater needs to heat a liter of water, the maximum amount of boilers that can be heated within a month based on their liter capacity was calculated. This was than compared with the amount of boilers a household would need within a month to provide them with enough hot water for showering, given the number of showers per person. Families with 10 individuals who on average shower 3 times a week needed approximately 31 more boilers per month to satisfy their showering needs. In other words, given the inputs that was gathered from households, on average, families with 10 individuals who shower 3 times a week need 31 more tanks of boiler than they can realistically heat during a month. It is important to mention that the same trend applied to a number of other families. On the other hand, households that have a positive amount of boilers may increase the frequency of showers that could be taken, or end up using the surplus for other activities within the household. Based on LSLP data it is more likely that households which in accordance with the calculations need more boilers per week to satisfy their needs, on average, take shorter showers, spending less than the European average. This is one of the factors that emphasis the difference of Kosovars well-being compared to the rest Europe. 4.6.2.

Household Appliances

Household appliances in the survey are categorized into several groups: utility room appliances,

kitchen

appliances,

entertainment

appliances

and

information

and

communications technology appliances. In order to calculate the electricity consumed by these appliances (in kWh terms) during a month the following formula was used. If the respondents indicated use in terms of daily use, the usage amount was multiplied by 7 for the following equation. (3) Households were asked the gender and age of the household members that participate in different activities. Culturally, Kosovo is still very traditional and they put a significant weight on having a clean and presentable house. These activities are dominantly done by women in the household (there were only 6 that indicated both genders participated). Women of all ages participate in the cleaning activities, with no age demographic presenting greater than another.

4.6.2.1.

Utility Room

Utility room appliances are washing machines, vacuum cleaners and irons. The ages of these appliances vary with most of the utility room appliances being newer than 5 years. The appliances that are older than 10 years are likely not as efficient, though very few households had appliances that were older than the conflict for the utility room.

Washing Machine

Appliance

0-5 6 – 10

Vacuum

11-15 16 – 20 Iron

20 +

0

100

200

300

400

500

600

700

Frequency

Figure 56. Age of Utility Appliances As shown in Figure 56, vacuum cleaners age range is quite diversified. The age of the irons, vacuum cleaners owned by these households ranges from less than 1 year old to 30 years old. The mean age of vacuum cleaners is 6.05 years. For Irons, 104 households or 19 percent of respondents that knew the purchase date, had purchased it within the last year. The large discrepancies in age terms are important indicators in explaining the difference in energy consumption. When looking at the age of these appliances with respect to the age of the home, as it would be assumed that many of these appliances would be purchased near the time of construction year. About 25% (n=151) of the sample purchased their washing machines around the time that the house was constructed; for vacuums 22% (n=133) and 20% (n=121) for irons, for purchasing within 4 years of the home’s construction. The consistent replacement rates for washing machines can be due in part to the significant amount of use the appliances fall under. Additionally, the households that have inconsistent quality in their energy service may also have higher rates of replacement. The drop off at 15 years 12/3/15

Trends and Potentials for LPLS Households- Draft

P.100

post construction is due to the conflict, though there are some appliances that remain from Yugoslavia. Table 32. Turnover rate for Utility Appliances by Year of Home Construction (n=605)

Same or less than 5 years post-construction 5-10 years post-construction 10-15 years post-construction 15+ years post-construction Kept from previous residence

Washing machines 25% 23% 24% 10% 5%

Vacuums

Iron

22% 16% 23% 10% 7%

20% 15% 20% 11% 8%

Most homes own at least one vacuum cleaner, 94.2 percent (570 households) had a vacuum cleaner, 1 percent owned two vacuum cleaners (6 households) meanwhile, 29 households or 4.8 percent either did not answer the question or did not own one. Based on the information obtained from the respondents, on average households use vacuum 4.5 hours per week. The largest portion of households, 120 (20%), use the vacuum cleaner 7 hours per week (Figure 57). Moreover, four respondents declared that they do not use their vacuum cleaner. Although 576 households had at least a vacuum cleaner only 294 were able to identify the wattage capacity. The wattage of vacuum cleaners owned by households ranges from 60W to 5.8 kW, therefore, the mean average is 1.67 kW.


140 120 100 80 60 40 20 .00 1.00 1.17 1.20 1.40 1.75 2.10 2.30 2.33 3.00 3.50 4.50 5.50 6.50 8.00 9.50 10.50 14.00 21.00 30.00

0

Usage per Week (in hours)

Figure 57. Vacuum cleaner usage 12/3/15


P.101

Irons are commonly found in households. From the survey, 94.9 percent or 574 households own at least one iron. Moreover, nine households own two irons, three households own three irons and only one household owns five irons. Households use their irons range from 0 to 8 hours per week, with an outlier at 25 hours. Five respondents that owned at least an iron declared that they do not use it. All in all, on average these households spend 2.43 hours per week ironing. Only 111 households answered the question about the wattage of their irons. 26 of these households, the largest portion, declared that their iron is large 2.0 kW. Meanwhile, the average iron wattage was 1.69 kW. Based on the amount of hours a vacuum cleaner is used per week and the vacuum cleaners’ wattage capacity, vacuuming uses on average 27 kWh per month. For ironing, on average households interviewed spend 10.4 kWh per month. Resulting in the average consumption of energy towards households’ utility appliances (excluding washing machines) is on average, 37.4 kWh per month. Washing Machines Washing machines are owned by 586 households or 96.9 percent of the respondents, 12 families owned more than one washing machine. For some households, despite owning at least one washing machine, do not use it. These respondents indicated that the costs associated with the usage of this appliance were high and laundered their clothes by hand, and boiled the water on the stove. Washing machines under 5 years may be more energy efficient, and 42.4% (n=256) of the sample own a machine that is less than 5 years old (Table 33). Older washing machines usually tend to have more wattage capacity than new ones. Most of washing machines with a wattage capacity of 5 kW are older than 5 years and older washing machines consumed significantly more water. Average consumption for top-loading washing machines older than 20 years, consume approximately 150 liters per cycle, whereas, today the average is 50 liters per cycle (efficient machines can run on about 35 liters/cycle). For front loading machines, in 1990 averaged 13.6 liters/kg to 7.2 liters/kg in 1997, and has not changed much since70.

70

European Commission (DG ENV), “Study on Water Performance of Buildings.”

12/3/15


P.102

Table 33 Age of Washing Machines Number of Washing Machines 244 112 140 24 4

Age (years) 20

Furthermore, other important factors that affect the amount of electricity consumed are the number of loads and the Celsius degrees used to launder clothes. As computed from the inputs of households that were surveyed most of the respondents (41.5%) wash their white clothes once per week, meanwhile, 36.3 percent of households wash their white clothes twice per week (Figure 58). The rest use their launder more than twice per week so as to wash their white clothes. On average households launder their white clothes 1.9 per week. The average temperature for white laundry is 90 to 95 degrees, with only 3% of the sample using colder temperatures.

250

Number of Homes

200 150 100 50

50.00

14.00

10.00

8.00

7.00

6.50

6.00

5.50

5.00

4.50

4.00

3.50

3.00

2.50

2.00

1.50

0.5

1.00

.00

0

Times per Week

Figure 58. Laundering Frequency On the other hand, only 16 percent of households wash their colored clothes once per week and an additional 19 percent wash their colored clothes twice per week (Figure 58). On average, households launder their colored clothes 3.06 times per week. Some 12/3/15


P.103

households declared that they wash their colored clothes 14 times a week, which means that on average they use their washing machines twice a day for colored clothes only. The average temperature for colored laundry is between 40 and 60 Celsius. When asked what the average weekly usage of the washing machine was, regardless of color or temperature, was 10.98 hours. The higher the number of laundries per week the higher the amount of hours a washing machine is used (Figure 59). Respondents were asked the brand of washing machines (Table 34). As a consequence the top four brands of washing machines are Gorenje (n: 331), Beko (n: 92), Bosch (n: 17) and Siemens (n: 14).

Figure 59. Washing Machine Usage and Frequency of Loads Table 34 Brand of Washing Machines Brand Gorenje Beko Bosch Siemens

4.6.2.2.

Number 331 92 17 14

Kitchen Appliances

Home appliances in general are responsible for a large portion of energy consumption. In particular kitchen appliances such as fridge-freezer, refrigerator, chest freezer, electric 12/3/15


P.104

oven or hub and microwaves are considered to be large consumers of energy71. Appliances like fridge, fridge-freezers or chest-freezers are a necessity for a family, but are considered to passively consume electricity (versus heating water). 4.6.2.3.

Cooling Storage Data LSLP shows that 57.2 percent or 346 households own at least one fridge. Only

nine households own 2 fridges meanwhile, 259 households either did not answer or do not have a fridge. Since fridges are used to conserve food or other products it is normal to leave it on 24 hours a day, but the energy footprint cycles when the fridge is cooling. Consumers typically do not unplug or discontinue the use of the refrigerators; however, 3.3 percent or 11 households admitted that they use the fridge only 12 hours per day. And 1.8 percent or 6 households said that they used the fridge 9 hours or less per day. From the households surveyed only 245 declared that they owned a fridge-freezer combination, the rest either did not own one or refused to answer. The vast majority of these households owned only one fridge-freezer however, five households owned two fridge-freezers and only one household owned five fridge-freezers. Most homes do not disconnect their fridges, except for 6 homes, which used their fridges between 9 – 14 hours per day. Nonetheless, two households despite owning at least one fridge-freezer did not use it. One strategy for saving food is the chest freezer, 447 households (73.89 percent) own at least one. Seven respondents that admitted the ownership of two chest-freezers and one respondent indicated that the home contains three chest-freezers. As expected most households do not unplug their freezers. Though, 14 respondents declared that they use their chest-freezer 10-15 hours a day. Only one of the households, despite owning a chest freezer, did not use it. Cold storage is a fundamental household appliance. When looking at the age range of these appliances in relation to the age of the home, very few of these appliances made it past the conflict years. However, when considering the turnover rate, the appliances are replaced at a rate of around 25% every 5 years. The lifespan of an appliance depends on many factors, including use, environment, electricity quality, etc., however, there are some rule of thumb metrics for these appliances. Freezers are expected to last 11 years,

71

“Appliance Energy Usage Guide.”

12/3/15


P.105

refrigerators from 9-13 years72. When considering the electricity quality on the age of the appliance, the households in the low-service area had no difference in replacement rates than those in the low payment areas. Anecdotally, households had indicated that they felt that there appliances needed to be replaced more frequently and that surges in the system had burnt out their appliances. Table 35. Turnover rate for Cooling Appliances by Year of Home Construction (n=605)


Fridge Freezer 22% 22% 25% 12% 4%

Fridge 23% 23% 24% 11% 4%

Chest Freezer 25% 25% 21% 13% 7%

Chest-Freezer 0-5 6 – 10 Fridge

11-15 16 – 20 20 +

Fridge-Freezer

0

100

200

300

400

500

Figure 60. Age of Large Kitchen Appliances As shown in Figure 60 the age of large kitchen appliance used by these households varies from less than a year to 20 years, with the largest portion of fridges, 43.4 percent, not older than 5 years. Of those surveyed, 20 percent, of households own fridge-freezers

David Seiders et al., “National Association of Home Builders/ Bank of America Home Equity Study of Life Expectancy of Home Components.” 72

12/3/15


P.106

that are in between 15-20 years old. When asked about the age of their chest freezers, 140 respondents said that they own chest freezers which are not older than 5 years; 147 households have chest freezers that are 5-10 years old and 112 had 10-15 years old (Figure 60). Some households (n: 36) noted that they own a chest freezer which is older than 15 years. The oldest chest-freezer owned by a household is 34 years old. The 25.4 percent of households own fridges older than 10 years are likely to be in inefficient in the use of electricity. These older appliances are likely more significant consumers of electricity. The average wattage of these appliances based on their respondents is 1.08 kW the range varies from 50 W to 2.8 kW. Out of 81 households that knew how many watts their fridge, 14.8 percent responded 2.0 kW; 42 households had fridges with wattages of 50 and 750W. Chest freezer wattage capacity varies from 66W to 8kW. It is important to emphasize that 66W is not the typical chest freezer size and should be considered an outlier. The mean average of the chest freezers wattage is 1.49 kW. When asked about the brand of fridge-freezers top 5 brands were Gorenje, Beko, Bira, Samsung and Bosch (Table 36). The 279 respondents that were able to identify the brand of their fridge, top 5 brands are Gorenje, Beko, Bira and Siemens For chest freezers, the top 5 brands are Gorenje, Beko, Bira, Bosch and Venus. Gorenje appears to have a significant market share for many appliances throughout the survey. It may be useful to train employees at the point of sale the benefits of using more efficient appliances. Table 36 Brand of Fridge-freezers Brand Gorenje Beko Bira Samsung Bosch Siemens Venus

4.6.2.4. 12/3/15

Number of Fridgefreezers 89 43 6 4 4

Number of Fridges 124 50 14

Number of Chestfreezers 161 65 4

7 3

4 4

Cooking Trends and Potentials for LPLS Households- Draft

P.107

Cooking is a task in the household that is again relegated to the women of the household, as only 3 households indicated that both genders participated in the cooking efforts. The options for cooking depend on the energy source that the household has available. Meals in the surveyed areas are likely to be cooked on the wood cook stoves, while heating the house. However, most homes have at least one electric appliance used for the preparation of food. Electric ovens & hubs, microwaves and table-top cookers are very likely to be part of a household kitchen, and are likely used in the summer months, for quick preparations or foods that need an oven. Electric ovens according to the answers of the respondents are present in 212 households; two of these households own two electric ovens. Anecdotally, a couple of households responded that the oven was purchased as a means to complete the kitchen appliances. Households were asked the number of hours that they used their electric oven per week. Only 176 respondents answered, out of which 40.6 percent or 71 households declared that they did not use their electric oven (Figure 61). Two households may be outliers which responded that they used their electric oven more 14-15 hours per day. The low hours of usage can likely be caused by households substituting out electric cook for their wood stoves. 80 70 60 50 40 30 20 10 0

Electric Oven

Electric Hob

Table Top cooker

Microwave

Figure 61. Usage of Electric Cooking Devices 12/3/15


P.108

Another appliance that is owned by a significant portion of households is the electric hob. These single hot plate devices are easy and quick to use. Of the sample, 89 households own one electric hob and one household owns two electric hobs. However, a large portion of households do not use their electric hobs (Figure 61). From 68 households that answered this question, 31 households (45.6 percent) declared that they do not use their electric hobs. Alternatively, there are 37 households who use their electric hob in daily basis. Therefore, the daily usage of electric hobs ranges from 0 to 3.5 hours with an average use per day of around 0.8 hours.

Table Top Cooker

0-5

Microwave

6 – 10 11-15 Electric Hob

16 – 20 20 +

Electric Oven

0

50

100

150

200

250

Figure 62. Age of Cooking Appliances In regards to age, six of respondents that owned at least one electric oven did not know how old it was. A large number of households (n: 85), indicated that their oven is not older than 5 years; 62 households had ovens 5-10 years old. and 45 households have ovens that are 15-20 years old (Figure 62). There are also some households (n: 14) that own ovens which are older than 20 years. Hence, the ovens age range varies from less than a year to 30 years. Only 30 households knew how many watts their oven has. As a result, the wattage capacity of electric ovens ranges from 35 to 9.4kW and the mean wattage of this appliance according to the calculation is 2.8 kW. Subsequently, on average, a household because of electric ovens uses 21.5 kWh per month. 12/3/15


P.109

As shown in Figure 62 the age of electric hobs ranges from less than a year to 30 years. 38 households own electric hobs that are 0-5 years old and 25 households have electric hobs that are old between 6 and 10 years. Furthermore, a small amount of households (n: 7) declared that their electric hob is older than 20 years. Only 14 households knew how many watts is their electric oven. As such, the electric oven wattage varies from 700W to 4 kW and the mean is 2.4 kW. Based on the information provided by the respondents, it resulted that on average these households spend 14.4 kWh per month by using their electric hob. Microwaves are part of households’ kitchen appliances too, though only 57 own a microwave. Of these, when asked about the weekly usage of their microwaves 12 households declared that they did not use it all (Figure 61). The weekly usage range in hour terms varies from 0 to 3.5 hours per week. Most of the microwaves are not older than 5 years (Figure 62). The oldest microwave owned by a household is 20 years old. The wattage of the microwaves that are owned by these households varies from 800 to 1.2 kW. The average household consumes 2.0 kWh per month by using their microwaves. For the table-top cooker, only 31 households (5.1%) said that they own this appliance. The usage varies from 0 to 6 hours and four households said that they did not use their table-top cooker. The age of this appliance varies from 1 to 25 years. Only four households could identify the wattage capacity of their appliance. Hence, the mean wattage of their table-top cookers is 2.5 kW. On average, these households consume 70.97 kWh per month by using their table-top cookers. In terms of appliance rate turnovers, given the light usage of these alternative cooking appliances the rates are much lower than for the other electronic appliances. The electric oven rates are interesting, especially in regards to the rates for the previous appliances (Table 37). The pattern suggests that the appliances in the household are bought new every 5 years at the same rate of houses.

12/3/15


P.110

Table 37. Turnover rate for Cooking Appliances by Year of Home Construction (n=605)


4.6.3.

Electric Oven 23% 23% 23% 14% 5%

Electric Hob 18% 30% 19% 13% 6%

Microwave 15% 19% 36% 13% 0%

Entertainment Appliances

The household entertainment appliances covered in the survey are Televisions (tube TV, LCD/plasma), digital receivers and radios. One of the differences between older televisions and the LCD/Plasma televisions are the build-in power saving functions that automatically shut-off the set based on time when no one is watching. Additionally, the television and digital receivers are used in tandem, though the receiver is a passive device that consumes regardless of use, unless turned off or unplugged. All households own at least one TV, whether tube or LCD/Plasma; 17 households own 2 TVs and three households own three TVs. For LCDs/Plasmas, 283 households own 1, while four households own two. The usage of TVs by households per day ranges from 0 to 20 hours (Figure 63). The average usage of TVs is 6.9 hours per day whereas; the consumption for the LCDs/Plasmas is slightly higher, at 7.88 hours per day.

12/3/15


P.111

40 35 Frequency

30 25 20

TV

15

LCD/Plasma

10

Digital Receiver

5 0

Hours per Day

Figure 63 Usage of TV, LCD/Plasmas, Digital Receivers 300 250 200 TV 150

LCD/Plasma Digital Receivers

100 50 0 0-5

6 – 10

11-15

16 – 20

20 +

Figure 64 Age of TV, LCD/Plasmas, Digital Receivers The largest portion of households (n: 59) own TVs that are 10 years old (Figure 64), which is not entirely unsurprising given the introduction of cheaper LCD/Plasma screens in the last decade. This can be seen in the proportion of households that own a new LCD 82.23 percent own a LCD/Plasma that is less than 4 years old. We asked households the age of their digital receivers to get an idea of how long they have had access to cable services, 70 percent of households noted that their digital receiver is not older than 2 years. 12/3/15


P.112

From the sample, 116 households were able to give us the wattage of their TV sets. TV’s wattage ranges from 15 to 600 W. Furthermore, only 137 respondents knew the wattage of their LCDs/Plasmas which ranges from 24 to 380 W. The wattage for the digital receivers range from 20 to 200 W. The newer sets, although there may be some skepticism on investing in a new set, the LCDs are more efficient in consumption. Though likely this effect is lost with the rebound effect of watching more TV, as the average households consumes 41 kWh with the older sets and 44 kWh using their LCD/Plasma per month. Add to this the amount of energy that is spent on the digital receivers which on average is spends 22 kWh per month. This means that watching TV is about 66 kWh per month, an amount that is likely spread over all of the tariff blocks, and mostly during the day. Radios were considered in the study given the remoteness of some of the villages and the likelihood that there would be some use for news and music. Only 61 households admitted to having a radio and two households said that they own two radios but the usage of radios is very limited, with almost a third never using them. The average usage of radio per day is 1.5 hours. Radios owned by these households are relatively old as the majority is older than 10 years. From the responses of 13 respondents it can be derived that the average wattage capacity of radios is 64.23 W. On average because of using a radio consumes 1.1 kWh per month. 4.6.4.

Computers

The survey covered communication/entertainment devices, which included computers (desktops, laptops) and cell phones. These devices will also likely use the digital receiver boxes, but we did not include a question on modems. Roughly half of (n: 302) households own at least one desktop (PC) and 173 households own at least one laptop. The shift from desktops to laptops it seems to have started 5 years ago, though there are still some homes that are purchasing desktops. Nevertheless, 284 households were able to remember the year that they purchased the computer and 163 respondents knew when they purchased their laptops. Of these households 45 owned laptops that were 2 years old, the average age for desktops was 5 years. Additionally, the daily usage range of desktop and laptops ranges from 0 to 24 hours 12/3/15


P.113

per day. Hence, the daily average usage of desktops amounts to 5.24 hours while the perday average usage of laptops is 3.63 hours. In addition, respondents were asked the brand of their desktops and laptops, too. As a result, the ranking for the top two known brands of both these appliances were Dell and HP. According to respondents, 101 households had Dell desktops and 35 own Dell laptops.

50 45 40 35 30 25 20 15 10 5 0

Laptop Desktop