Disclaimer

The Energy Regulation Board (ERB), consistent with its’ mandate of regulating the energy sector
in Zambia, does carry out specialized studies that encourage the exchange of ideas about energy
regulatory impact analysis and development issues in general. This particular study was jointly
undertaken with the Central Statistical Office (CSO). However, the findings, interpretations and
conclusions expressed in this paper are entirely those of the authors. They do not necessarily
represent the views of the ERB, CSO, allied institutions or the Government. Thus, the study carries
the names of authors and should be cited accordingly.

The study team members for the Load shedding study comprised the following: Mr.
Alfred Mwila, Director – Economic Regulation, ERB; Mr. Goodson Sinyenga, Deputy
Director – Economics and Financial Statistics, CSO; Mr. Simweemba Buumba, Senior
Manager-Research and Pricing, ERB; Mr. Rodgers Muyangwa, Manager-Electricity;
ERB, Mr. Namakando Mukelabai, Statistician, ERB; Mr. Cletus Sikwanda, Economist-
Research, ERB; Mr. Besa Chimbaka, Economic Analyst-Electricity, ERB; Mr. Gerson
Banda, Senior Statistician, CSO; and Ms Chenela Nkowani, Programmer, CSO.

2
 ENERGY REGULATION BOARD

IMPACT OF LOAD SHEDDING ON

SMALL SCALE ENTERPRISES

by

Alfred Mwila, Goodson Sinyenga, Simweemba Buumba, Rodgers Muyangwa,

Namakando Mukelabai, Cletus Sikwanda, Besa Chimbaka, Gerson Banda;
and Chenela Nkowani.

Working Paper 1
June 2017

Energy Regulation Board (ERB)

Plot 9330, Off Alick Nhata Road
Lusaka, Zambia
Downloadable at [Link]

i
 Table of Contents
Abbreviations ............................................................................................................................................... iv

Executive Summary........................................................................................................................................ v

Chapter 1: Introduction.................................................................................................................................1

1.1 Introduction....................................................................................................................................1

1.2 Problem statement and justification......................................................................................6

1.3 Study Objectives............................................................................................................................7

1.3.1 General Objective..................................................................................................................... 7

1.3.2 Specific objectives....................................................................................................................7

1.4 Structure of the paper..................................................................................................................7

Chapter 2: Theoretical Framework............................................................................................................9

2.1 Cost of unsaved energy ............................................................................................................10

2.2 Estimating the cost of power rationing...............................................................................10

CHAPTER 3: METHODOLOGY....................................................................................................................15

3.1 Approach and Methodology...................................................................................................15

3.2 Sample Survey Coverage and Target population............................................................15

3.3 Sampling Design..........................................................................................................................16

3.3.1 Sampling frame............................................................................................................................16

3.3.2 Sample Size Determination and Allocation.......................................................................16

3.3.3 Sample weights and Sampling...............................................................................................16

3.4 Data analysis and techniques..................................................................................................17

3.5 Limitations.....................................................................................................................................17

CHAPTER 4: RESULTS AND DISCUSSION OF FINDINGS....................................................................19

4.1 Distribution of small scale enterprises by sector.............................................................19

4.2 Number of workers employed by small scale enterprises............................................21

4.3 Wage bill.........................................................................................................................................22

4.4 Business working hours............................................................................................................22

ii
 4.5 Load shedding Experience.......................................................................................................23

4.6 Electricity expenses by small scale enterprises................................................................25

4.7 Business Annual Turnover........................................................................................................26

4.8 The impact of load shedding on turnover..........................................................................27

4.9 The Impact of load shedding on labour costs..................................................................28

4.10 Equipment damage and maintenance attributed to load shedding.......................30

4.11 The cost of restarting operations as a result of load shedding...................................31

4.12 Demand side management strategies................................................................................32

4.12.1 Use of generators....................................................................................................................32

4.12.2 Use of uninterruptible power supply...............................................................................33

4.12.3 Surge protectors......................................................................................................................33

4.12.4 Back up data systems.............................................................................................................34

4.12.5 Use of back up batteries.......................................................................................................34

4.12.6 Security enhancement..........................................................................................................35

4.12.7 Reduction in labour................................................................................................................35

4.12.8 Reduction in the working hours........................................................................................36

4.12.9 Relocation of business...........................................................................................................36

4.12.10 Shutdown of business operations................................................................................37

4.12.11 Change of operations hours...........................................................................................38

4.12.12 Other measures taken.......................................................................................................38

4.13 Challenges due to load shedding..........................................................................................39

4.14 The impact of load shedding on small scale enterprises..............................................40

Chapter 5: Conclusion and recommendations...................................................................................42

References ..............................................................................................................................................46

iii
 Abbreviations
CSO Central Statistical Office

ERB Energy Regulation Board

GDP Gross Domestic Product

GWh Giga-Watt hour (1,000 MWh)

IPP Independent Power Producer

SBR Statistical Business Register

SMEs Small and Medium Sized Enterprises

ZESCO ZESCO Limited

Source: SMEs Website, Zambia

iv
 Executive Summary

Background

The Electricity Supply Industry (ESI) in Zambia is dominated by hydro generation which in
2015 accounted for 94.1% of national installed capacity. The balance of 5.9% was from
alternative sources such as Diesel, Heavy Fuel Oil (HFO) and Solar Photovoltaic (PV)
generation plants. In 2015, Zambia experienced a drastic reduction in electricity supply
which was attributed to the reduced generation by ZESCO Limited (ZESCO) due to the low
water levels in the reserves caused by poor rainfall in the 2014/15 rainy season. The
power deficit in 2015 ranged from 560 to 1000 MW. By July 2015, ZESCO had increased the
extent of load shedding to at least eight (8) hours a day for the majority of its household,
commercial and industrial consumers. One of the measures of load management undertaken
by ZESCO was load shedding. The load shedding affected the most business operations and
financial viability. From literature, it has been investigated that small enterprises are the
most likely to be adversely affected by measures such as load shedding. This is because,
small enterprises are less resilient and most of them are not insured or have limited capacity
to invest in alternative energy sources (Kazungu, Moshi, & Mchopa, 2014). Given the
importance of small enterprises in the economy, it is critical that the impact of load shedding
is studied and understood. For example, according to Nuwagaba, (2015), considering
the data on Small and Medium Sized Enterprises (SMEs) for the period 1993-2006, SMEs
had created total employment of 214,527 in different sectors of the economy. Agriculture
sector provided 36.7 percent followed by manufacturing with 34.3 percent.

Objective of the study

The objective of this study was to ascertain the extent of the impact of load shedding on
small enterprise business operations and financial performance of smallscale enterprises in
2015.

Methodology

The generic approach to estimating the impact of load shedding or unserved energy is the
Cost of Unserved Energy (COUE). The COUE is defined as the value in monetary terms (e.g.
Kwacha per kWh) that is placed on a unit of electricity not supplied.

This study used the Direct Assessment Method (DAM), a derivative of the COUE, to estimate
the impact. The DAM estimates the cost of power outages by allowing electricity consumers
to express their losses in monetary terms (Kaseke & Hosking, 2012). The approach is based
on the principle that the lost production, materials and time in each productive sector, or
lost goods during an outage (load shedding), can be estimated directly, and this can be
aggregated to a total (ibid, 2012). The approach relies on the individual respondent’s
self-assessment method of valuing the cost of electricity outage.

The scope of the study was limited to four (4) cities, namely, Lusaka, Kitwe, Ndola, and
Livingstone. The four districts were purposively sampled owing to their relatively higher

v
concentration of economic activities. The study employed both qualitative and quantitative
techniques to select samples from the Central Statistical Office (CSO) sampling frame.
The small enterprises sample was limited to establishments whose annual turnover did
not exceed ZMW 250,000.00 as per CSO definition. Sampling weights were used to
correct for differential representation of the sample due to the disproportionate allocation
of the sample and this made it possible to make reference to the rest of the population in
the survey areas.

General Characteristics of small enterprises

The general demographic and characteristics of the sample were as follows:

• Sixty-four (64) percent of the enterprises were formalized by a way of registration with
one of the one of the local authorities such as Zambia Revenue Authority (ZRA), National
Pension Authority (NAPSA), The Patents and Companies Registration Authority (PACRA)
and the Local Council.

• Twenty-Seven (27) percent of enterprises had social security schemes or did make
contributions on behalf of their employees to NAPSA, Local Superannuation Fund
(LASF), Public Service Pension Fund (PSPF) and some any other social security system.

• The period of establishment for small enterprises ranged from 1922 to 2016.

• In 2014 and 2015, only 17.8% and 18.2% of the small enterprises were insured
respectively.

• The highest ranking operational constraint was electricity which was reported by 35.8%
of the enterprises in the survey. This was followed by Finance at 27.7%.

• In terms of the legal status of the establishment, the majority (62%) were individual
proprietorship followed by private limited companies (23.1%) and partnerships (7.2%).

• The wholesale and retail trade had the highest number of small scale enterprises at 7,533
(48.9%), followed by accommodation and food services and other services activities
at 2,771 (18.0%) and 1,868 (12.1%) respectively.

• In the survey areas, small scale enterprises provided employment to a total of 174,028
employees in 2015 and on average 11 employees per enterprise. In terms of sex,
70,644 (41%) were female and 103,383 (59%) were male.

• The annual wage bill for the establishments ranged from K136 to K466,869.45 in 2015.
The average wage bill for each enterprise was K 13,474.33

• The average number of working hours per day for small scale enterprises was 11 hours.
Meanwhile, the average number of operating days per week was 6.

vi
• A total of 12,781 (83.1%) of the small scale enterprises, in the survey areas, experienced
load shedding in 2015.

• Overall, on average the total number of hours of load shedding per day increased from
1 hour per day in 2014 to 4 hours per day in 2015, representing a percentage increase
of 300%.

• Load shedding schedules were not strictly followed by ZESCO as reported by 51%
of the enterprises. Meanwhile, the common source of information for load shedding
schedules was the Short Messaging System (SMS).

• A total of 2,125 (17%) indicated that they did not receive information on
ZESCO load shedding schedules.

• Electricity expenses ranged from K50 to K 14,083 per month in 2015. The
average electricity expenditure was K754.61.

• The majority (85%) of the small scale enterprises on average indicated that
electricity bills constituted up to 25 % of their total annual business expenses.

• The annual turnover, defined as total sales of the business in a year, ranged from
K 2,200 to K 10,999.99. The average annual turnover was K407,527.

Impact of Load Shedding

• The reported loss in turnover as a result of load shedding ranged from K0 to
K759,000. On average the reported loss in turnover was K 19,251.16.

• In the survey, 22.8% of establishments reported cases of idle labour while 9.8%
reported incurring overtime labour costs due to load shedding. Idle labour costs on
average ranged from K0 to K8,333.33. The average idle labour costs per firm were
K130.80. Equally, at firm level overtime labour costs on average ranged from K0 to
K5,000. The average overtime labour costs per firm was K30.49

• In the survey, 29.9% reported damaged equipment due to load shedding.

Furthermore, only 11.7% reported that such equipment was insured. The cost of
damaged equipment ranged from K2 to K56,000. The average cost of damaged
equipment was K3,112.80.

• A total of K 3,754,390.00 was spent in the restarting of operations by small scale

enterprises in 2015 in the four districts as a result of load shedding.

Measure to mitigate load shedding

• In the study, 55% reported employing strategies to mitigate against load shedding.

vii
 However, there was no reported shutdown of operations and very few reallocations
of businesses (0.2%).

• A total of 3,511 (29.9%) of small scale enterprises used generators as an

alternative source of energy, while 8,251 (70.1%) did not.

• Only 276 (2.4%) used uninterruptible power supply (UPS) as an alternative

energy source to mitigate against the impact of load shedding.

• Only 246 (2.1%) indicated that they used surge protectors in the four districts.

• Only 813 (6.9%) of the small scale enterprises invested in back up data systems.

• Only 803 (6.9%) used back up batteries in order to mitigate against the loss of power.

• In the survey, 803 (6.9%) small enterprises reported that they had enhanced their
security features.

• A total of 1,013 (8%) out of 12,452 small scale enterprises indicated that they
reduced the labour hours compared to 11,439 (92%) who did not.

• A total of 876 (7.6%) indicated that they reduced business working hours due to
load shedding.

• A total of 808 (6.5%) of the small scale enterprises indicated that they changed their
business operating hours in order to accommodate the load shedding schedule.

Economic impact of load shedding

The study established that a total cost of K623,871,514.50 was incurred as a result of load
shedding by small scale enterprises translating into US$ 0.95/kWhlos (kilowatt hour lost).
In terms of kilowatt hour lost per district, Kitwe district had the highest loss at US$ 1.94/
kWhlos, followed by Lusaka at US$ 0.97/kWhlos. Livingstone district was third at US$ 0.53/
kWhlos, while Ndola was the lowest at US$ 0.51/kWhlos.

Conclusion
This study has established that the incidence of load shedding in 2015 led to adverse
disruptions in the operations of most small enterprises in the survey areas. Furthermore,
most small enterprises had inadequate response strategies as they could not use alternative
sources of energy. Most small enterprises resorted to reducing their work outputs resulting
in reduced turnover whilst incurring additional costs such as idle labor and overtime. Some
enterprises suffered losses due to equipment damage and high replacement costs.
The study estimate of US$ 0.95/kWh for each unsupplied electricity unit confirms the
proposition that small enterprises were adversely affected by load shedding and that there
is an inverse relationship between load shedding and small enterprise productivity as
well as general business performance.

viii
 Chapter 1

Introduction

1.0 Introduction

Zambia is a landlocked country in Southern Africa with a total surface area of 752,618 square
kilometers and had a population of approximately 15,473,905 in 2015 (Central Intelligence
Agency, 2017). Zambia has had one of the world’s fastest growing economies for the past
ten years, with real Gross Domestic Product (GDP) growth averaging roughly 6.7% per
annum, though growth slowed in 2015 to 2.9%, due to falling copper prices, reduced power
generation, and depreciation of the kwacha. Zambia’s lack of economic diversification and
dependency on copper as its sole major export makes it vulnerable to fluctuations in the
world commodities market and prices turned downward in 2015 due to declining demand
from China.

According to the Central Statistical Office (2016), by 2015, GDP at current prices was estimated
at K183,381.1 million compared to K167, 052.5 million in 2014. The results show that the
Wholesale and retail trade industry had the highest contribution of 22 percent to GDP in
both years. This was followed by Mining and quarrying industry at 14.6 percent in 2014 and
12.7 percent in 2015. The share of Agriculture, forestry and fishing reduced from 6.8 percent
in 2014 to 5.0 percent in 2015. The contribution of electricity generation to GDP increased to
4% in 2015 compared to 3% in 2014.

In 2015, the economy experienced a deterioration in its terms of trade owing to a decline
in exports. The Kwacha depreciated significantly against major international currencies.
The economy witnessed double digit inflation, for the first time since 2010. The Kwacha
depreciated by 25.0%, from K6.47/US$ to K8.09/US$ between January and August 2015, while
between September and December 2015, the Kwacha depreciated by 6.3%, from K10.20/US$
to K10.84/US$. Notably, the Kwacha depreciated significantly by 48.57% between August
and October 2015, owing to the highest trade imbalance recorded in October, 2015 and
the collapse of the copper prices which is Zambia’s main export. Inflation averaged 10.04%,
rising from 7.7% in January to 21.1% in December, mainly driven by the depreciation of the
exchange rate. In line with inflationary pressures, interest rates remained relatively high with
the commercial banks’ lending rates increasing to 23.9 percent at end of December 2015
from 20.5 percent at end of December 2014 (Ministry of Finance, 2016).

In 2015, the Electricity Supply Industry (ESI) in Zambia was dominated by hydro generation
which accounted for 2,269 MW (94.1%) of national installed capacity and the balance of 5.9
percent was from diesel (92 MW) , Heavy Fuel Oil (50 MW), and Solar Photovoltaic (0.06 MW)
generation plants. Figure 1 shows Zambia’s installed generation capacity by technology in
2015.

1
Solar Photovoltaic (0.06 MW) generation plants. Figure 1 shows Zambia’s installed
generation capacity by technology in 2015.

Figure1:
Figure 1: Zambia’s
Zambia’sinstalled
installedgeneration
generationcapacity
capacity – 2015
– 2015

Diesel Solar PV
3.8% 0.0%
Heavy Fuel oil
2.1%

Hydro
94.1%

The key players in the sector were ZESCO, a vertically integrated power Utility, which
2
generates, transmits, distributes and supplies electricity. It is a public utility, with the
Government of the Republic of Zambia being a sole shareholder. Other players included:

• The Copperbelt Energy Corporation (CEC) which operates and maintains a network
mainly comprising generation, transmission and distribution assets that supplies
power to Zambia’s mining companies based on the Copperbelt province.

• Lunsemfwa Hydro Power Company Limited an Independent Power Producer (IPP)

that supplies power solely to ZESCO with a total installed capacity of 56 MW
hydro power stations.

• Kariba North Bank Extension Power Corporation Limited, a wholly owned subsidiary
of ZESCO that owns and operates a 360 MW hydro power plant.

• North Western Energy Corporation Limited (NWEC) is licensed to distribute electricity

in the North-Western Province of Zambia. NWEC distributes electricity to non-mining
customers in Lumwana (Barrick), Kabitaka and Kalumbila sites. Power is supplied
by ZESCO at various substations established by NWEC.

• Ndola Energy Company Limited (NECL) an IPP that supplies power to its sole
customer, ZESCO, under a Power Purchase Agreement (PPA). The company

2
 owned and operated a 50 MW HFO power plant, which is planned to increase by a
further 55 MW HFO power plant beyond 2015.

• Zengamina Power Limited a private company that owns and operates an off-grid
mini hydro power plant with an installed capacity of 0.75 MW situated in Ikelenge,
North-Western Province.

The total generation sent out from both ZESCO and IPPs power plants declined by 7.0 percent
(1,013 GWh) in 2015. Electricity sent out reduced from 14,453 GWh in 2014 to 13,440 GWh
in 2015. The reduction in electricity generation was attributed to poor rainfall experienced
during the 2014/2015 rainy season which resulted in low water levels, thereby impacting
negatively on the capacity to generate power from hydro power plants.

In order to address the imbalance in electricity generation sent out, ZESCO undertook load
management measures which included load shedding. Load shedding is defined as an
intentionally engineered electrical power shutdown where electricity delivery is stopped
for non-overlapping periods of time over different parts of the distribution region1.
There are several factors that can cause load shedding besides insufficient generation
capacity. These factors include inadequate transmission and distribution infrastructure for
the delivery of sufficient power to the area where it is needed. The process is usually done
in stages and depending on the deficit, the utility company might decide to switch off
some segments of the customers during this process. Load shedding is a measure of last
resort to prevent the collapse of the entire power system. When the demand, or load, from
customers is greater than the available supply, the electricity system becomes unbalanced,
which can consequently result in country-wide power trips (a blackout) that could take days
2
to restore .

Particularly in 2015, ZESCO had increased the extent of load shedding from an average of
one (1) hour to between four (4) and eight (8) hours a day for the majority of its household,
commercial and industrial consumers. The power deficit in 2015 ranged from 560 to
1000 MW, and a load shedding schedule for different regions around the country was
developed by the Utility.

1
[Link]
2
[Link]

3
In order to mitigate against the power deficit, the Government instituted the following
measures amongst others:

i. Facilitation of the importation of emergency power from various sources within the
region;

ii. Announcement of a ban on local manufacturing and importation of incandescent

bulbs and inefficient lighting devices in January 2016 through Statutory Instrument
(SI) No.74 of 2016;

iii. The Government through the Industrial Development Corporation (IDC) in 2015,
commenced the procurement for the development of two solar power plants of 50
MW each to be awarded to two different developers; and

iv. Further, a new Lunzua power plant, owned by ZESCO and situated in Northern
Province, was constructed and commissioned with a rated capacity of 14.8 MW
adding to the existing capacity of 0.75 MW.

Electricity is a prerequisite for proper functioning of nearly all sub-sectors of the economy.
It is an essential service whose availability and quality determines success or failure of
development endeavors. This argument is valid particularly when we consider supply of
energy to small and large firms/businesses dealing with service provision and manufacturing,
where power is used as an input in the operations/production process rather than a final
consumption service. Hence, a temporary stoppage of power can lead to relative chaos.
While a loss of power in smaller scale settings may not be life threatening but can result
in lost data, missed deadlines, decrease in productivity or loss of revenue (Kazungu,
Moshi, & Mchopa, 2014).

Research on the effect of electricity power outage on Small and Medium Enterprises
(SMEs) in Ghana posited that, the electricity crises in the country costed SMEs over
US$686.4 million of annual sales. Based on previous research findings using a population of
over 4 million SMEs in Ghana with a sample size of 1,250, micro businesses were the most
affected by the electricity problems, recording a loss of around US$2.2 million daily,
which represented over 50% of their daily sales. The impact of power outages is dependent
on the firm’s ability to respond to any shocks, small scale enterprise have little room to
respond compared to medium scale firms (Solomon & Yao, 2015)

In a study carried out in Tanzania in 2014 using a survey research design, Kazungu,
Moshi and Mchopa found that SMEs experience various challenges with power rationing
being one of them. The study found that there was a strong positive correlation
between power rationing and decline in productivity. The study established productivity
loss was highest among SMEs that depended highly on electricity for their business
operations. Specifically, business declined between 50 percent and 60 percent for SMEs
dealing in photocopying and printing, stationery, hair dressing, barbershop and grain

4
milling (Kazungu, Moshi, & Mchopa, 2014). The occurrence of power rationing deprives SMEs
electricity for running their operations effectively and as a result, production is hampered
as there is no power to drive the business.

In the case of Zambia, as a result of long hours of load shedding, there was an outcry by
ZESCO’s customers concerning the negative impacts of load shedding on their routine and
core business activities. In particular, some businesses especially small ones, indicated
that they had to lay off workers while others had to close as they could not generate
sufficient revenues due to reduced production, to meet the business expenses. Furthermore,
some small scale enterprises complained of damaged materials and equipment. It was
therefore likely that such impacts would adversely affect the country’s GDP. Sing’andu
(2009) in a study to assess the impact of ZESCO’s power rationing on firm productivity
and profitability of selected manufacturing industries in Lusaka district, established that
power rationing eventually leads to a decline in production and consequently SMEs fail to
reach their projected sales volume. Reduced sales volume translates into reduced business
income because SMEs are unable to meet customer demand.

According to Nuwagaba (2015), SMEs are instrumental for the development of an economy
through, for example, employment creation, increased tax base for the country, and
improved incomes for the low earners among other benefits. Additionally, based on the
1996 baseline survey, SMEs employed 18 percent of the labour force of which 47% were
women in Zambia (ibid, 2015). Therefore, load shedding for such a strategic sector can
have devastating effects on the economy.

Firms suffer three kinds of damages in the case of an outage. First, they produce less, without
electricity, many production processes stop, some production is lost, for example unsaved
computer files, and it takes time to start up production again. Second, extra costs may be
incurred such as paying overtime pay to workers. Third, some goods and inputs may be
damaged, for example hot steel in a steel plant may cool down and have to be reheated. The
damage caused by an electricity interruption in a firm is equal to the value it would normally
have added during that period (Kaseke & Hosking, 2012).

There is no doubt that small scale enterprises are instrumental in the development of
the economy through employment creation amongst others. Additionally, they also
contribute to the treasury of the economy through tax payments. According to Andrew et
al., (2014) there were around 90 million micro, small and medium scale enterprises (MSMEs)
in developing countries and emerging markets and the density of formal MSMEs in low and
middle income countries is rising.

There is no consensus on the definition of an SME, as various countries have different

definitions depending on the phase of economic development and their prevailing social
conditions. Enterprises differ in their levels of capitalization, sales and employment. Hence,
definitions which employ measures of size (number of employees, turnover, profitability,

5
net worth, etc.) when applied to one country could lead to all firms being classified as
small, while the same size definition when applied to another country could lead to a
different result (Kanlisi , Amenga, Akomeah , Amoako , & Narh, 2014).

In this study, the definition adopted for a small scale enterprise was based on the
Central Statistical Office (CSO) classification of business enterprises, a small scale business
enterprise refers to a business whose annual turnover does not exceed ZMW 250,000.00.

In terms of the nature of business, most SMEs are engaged in the production of goods
and services with the primary objective of generating employment and income to persons
concerned. The range of products and services that most SMEs are involved in include
textile products, carpentry & other wood products, light engineering and metal fabrication,
food processing, leather products, handicrafts and ceramics. The services sector include
restaurants and food preparation, hair salons and barbershops, passenger and goods
transport, building construction, telecommunication services, business centre services and
cleaning services. The trading sector is largely concentrated in consumable products,
industrial products, and agricultural inputs and produce (Ministry of Commerce, Trade
and Industry, 2007). The business is characterised by the use of low technology, relying
largely on social networks and inter- firm cooperation, and are oriented towards the local
and less affluent segments of the market (Ibid, 2007).
1.1 Problem statement and justification

The persistent and long hours of load shedding experienced in 2015 by small and
medium enterprises in Zambia adversely affected their business operations and financial
viability. The problem of this study was to ascertain the extent to which load shedding
affected small scale enterprises in Zambia. The study is critical because small scale
enterprises are instrumental in the development of an economy given their contribution
to employment creation, increase tax base and improved incomes especially for low
income earners. Therefore, load shedding for such a strategic sector can have devastating
effects on the economy. Therefore, it becomes imperative to understand the financial and
operational impact of load shedding on small scale enterprises in Zambia, who are
presumed to be the most affected.

This study will undertake in-depth analysis of critical aspects that affect small scale enterprises
business operations affected by load shedding such as cost of material lost, the labour cost,
cost of equipment damage and maintenance and cost of restarting the business activities.

The study will also investigate the loss in turnover due to load shedding including the
different coping strategies put in place.

6
1.2 Study Objectives
1.2.1 General Objective

The overall objective of the study was to ascertain the operational and financial impact of
load shedding on small scale enterprises in Zambia during the load shedding experienced
in 2015.
1.2.2 Specific objectives
The specific objectives are as follows:
i. Ascertain the loss the loss in turnover and associated costs due to load shedding;
ii. Ascertain the cost of material lost due to load shedding;
iii. Ascertain the impact of load shedding on labour costs;

iv. Ascertain the cost of equipment damage and maintenance attributed to load
shedding;
v. Ascertain the cost of restarting operations as a result of load shedding; and
vi. Ascertain the measures put in place by enterprises to tackle load shedding.
1.3 Structure of the paper

This paper is structured as follows; chapter 1 provides the background, justification,

problem statement and objectives of the study. The rest of the paper shall proceed as follows:
Chapter 2 discusses the theoretical framework with regards to estimating the costs of
load shedding. Chapter 3 discusses the methodology employed in the study to estimate
the impact of load shedding on small scale enterprises business operations. Chapter 4
discusses the research findings while chapter 5 concludes and makes recommendations
based on the findings of the study.

7
 8
Source: SMES Poultry Website, Zambia
 Chapter 2

Theoretical Framework

2.1 Cost of unsaved energy

The generic approach used to estimate the cost of power outage is the estimation of the
cost of unsaved energy (COUE). COUE is the value of production lost for each unit of power
outage (Terry, 2001). It is also the monetary value placed on a unit of electricity not
supplied as a result of unplanned outages (Minnaar, 2015). It is estimated by:

GDP

COUE = ----------------------------------

Energy consumed

Using this approach, the Indaba Agricultural Policy Research Institute established that the
COUE in the agriculture sector in 2014 at ZMW 1.38/kWh, while for all sectors it was at
ZMW 15.53/kWh for Zambia. This implies that the agricultural sector was paying an implicit
price of ZMW 1.38 per unit of electricity. In 2015, based on an electricity shortfall of
2,100,000,000 kWh, the value of lost opportunity for all sectors was estimated to be ZMW
32,496,100,813 (that is, 18.8% of the GDP) in Zambia. For the agricultural sector, assuming
an 8.7% contribution to GDP in 2015, the estimated cost of the power shortfall in 2015
translated to ZMW 2,827,160,771 (1.6% of the GDP).

In 2015, Nyamazana (2015) estimated the COUE in Zambia for the years 2012 to 2015 as
depicted in table 1.

Table 1: Cost of unsaved energy for Zambia – 2012 to 2015

2012 2013 2014 20153

Total Electricity Cons (kWh Millions) 10,317 10,846 10,721 11,450

GDP (ZMW Millions) 128,370 144,722 165,901 183,381

GDP (US$ Millions) 24,939 26,821 27,066 21,249

COUE: ZMW/kWh 12.44 13.34 15.48 16.02

COUE: US$/kWh 0.194 0.185 0.163 1.856

3 2015 figures are author’s computations using average exchange rate of US$1 to ZMW 8.63

9
In 2015, the COUE was estimated at 1.86 US cents per kWh given the deficit of 2,100 GWh
and GDP of US$ 21,249,258,400.93. The loss was equivalent to US$ 3.9 billion or 18% of
GDP.

2.2 Estimating the cost of power rationing

At firm level, there are several approaches that have been used in literature to estimate
the cost of power rationing on different customers or sectors within the economy.
These approaches differ depending on the level of complexity and data requirements.
Some of these methods include the following Contingent valuation; Production function;
Captive generation method; and Direct Assessment Methodology (DAM).

2.2.1 Contingent valuation

According to (Samboko, et al., 2016) the contingent valuation approach is used

where consumers are asked to provide estimates of how much compensation they
would be willing to accept for a given period without power, or how much they are
willing to pay to avoid a power outage.

For example, a question could be phased as follows: If the incidence of outages

is reduced to half its present level, how much more would you be willing to pay on
your monthly electricity bill? An alternative approach would be to ask the following
question: If level of outages were to double, what reduction in your monthly electricity
bill would you consider to be fair? (Institute of Public Policy, 2013).

However, this method is prone to giving biased estimates as it is based on subjective

responses. It is likely that in response to the first question, the consumers understate their
willingness to pay for improved service, while they may overstate the compensation that
they would like to receive for deterioration in the reliability of supply. Using this
approach to determine the welfare costs of electricity outages in Uganda, Kateregga
(2009) found that the costs associated with load shedding varies according to the
incidences, particularly to the time during the day, morning or evening, and the duration
of the outage.

2.2.2 Production function

The production function approach achieves the same objective by providing estimates of
the input cost effect and the output loss from switching to alternative power sources. This
is usually done using panel data from firms on inputs and outputs (Samboko, et
al., 2016). The production function approach requires detailed data on individual firms,
however, that may not be easy to collect.

10
2.2.3 Captive generation

The captive generation method or the indirect method estimates the costs associated with
load shedding from the actions taken by consumers to mitigate outages by acquiring
generators or captive power units and diesel pumps. This method dates as far back as the
World War One when it was used by the US Navigation Army (1917) and was adopted by
British and other European countries in the 1930s as a way of consolidating their industry
production and estimating the negative effects of power outages (Nyasha , 2014).

Captive generation method is based on observed market behavior, for instance consumer’s
expenditures on generators and use of interruptible power supply contracts Firms are
assumed to be operating to maximize profits, while households are assumed to maximize
utility. A firm or household, faced with frequent power outages, will act to insure itself
against the damage caused, by acquiring backup generating units (ibid, 2014). The gain
from insurance against outages consists of the continued production or the continued
leisure that the self-generated electricity makes possible, and the avoided damage to
equipment that otherwise would have been caused by power outage (Opcit, 2014).
The expected gain from the marginal self-generation kilo watt hour (kWh) is also the
expected loss from the marginal kWh that comes as a result of an outage (Nyasha , 2014).

This method is easy to apply and can provide accurate estimates of the costs to firms as
data on the size of device generating units, the cost of the backup systems and the output
of the system, in the form of power (kWh) generated, can be easily traced to the suppliers
of the devices (Nyasha , 2014). The same information about the output can be traced to
the load that is powered by the device. The units required for these devices are known
internationally, e.g. the cooker consumes 60 AMPs on average and lights 10 AMPs (ibid,
2014).

However, critics argue that the use of the backup generation method to estimate cost
depends on whether the backup power supplies are for emergency or optional standby
(Caves et al. 1992). Where captive generation is used as (normal) emergency backup power,
the method may overestimate cost. On the other hand, Tiwari (2000) argues that power
outage costs are far greater than the backup generation costs, as there are indirect costs
other than direct costs that must still be added.

Further, the method assumes a perfectly competitive market for generators, risk neutrality,
and a production technology in which electricity enters smoothly. The existence of risk
aversion, externalities (which bring about environmental regulation), and technologies in
which relatively small generators, are of no use, would yield misleading estimates of the
marginal outage cost (Nyasha , 2014).

11
2.2.4 Direct Assessment Method

The Direct Assessment Method (DAM) is an economic appraisal tool that estimates the
cost of power outages by allowing electricity consumers to express their losses
in monetary terms (Kaseke & Hosking, 2012). The approach is based on the principle that
the lost production, materials and time in each productive sector, or lost goods during an
outage (load shedding), can be estimated directly, and this can be aggregated to a total
(ibid, 2012). The approach relies on the individual respondent’s self-assessment method of
valuing the cost of electricity outage.

In order to estimate the cost of load shedding by the DAM, it is important that total
value lost by consumers due to load shedding is ascertained by summing up all the
direct costs experienced during load shedding. The direct costs incurred by firms go
beyond production loss or output loss. In addition to output loss cost, other direct costs
such as materials destruction cost; in stock, labour cost; payment of idle labourers and cost
of overtime and bonuses to meet production and orders, damage to equipment cost, restart
cost, as well as time or opportunity cost per load shedding are part of the load shedding
cost. The total direct cost relationship is captured in the formula below:

TDCi = OLi + MCi + LCi + EDCi + MCi + RCi ………………………………………equation 1

Where: TDCi is the total direct cost for the ith consumer; OLi is cost of lost output; MDCi
is the material destruction cost; LCi is labour cost; EDCi is the equipment damage and MCi is
the maintenance cost as a result of load shedding; and RCi is restart cost.

From equation 1 cost per unit of electricity (kWh) lost can be estimated as:
TDCi
OCi = ---------------------------------------------------------
kWhlosi

Where: OCi is the cost per kWh lost and kWhlosi are the total units of electricity (kWh) lost
or unsupplied due to load shedding.

This method, however, has its own shortcomings. The DAM approach only measures
direct cost of production such as lost output, and not indirect cost such as inconvenience.
In addition, this method does not take into account the fact that foregone production might
be partially made up after the outage and as a result of this, gives an overestimation of

12
the cost of electricity outages. Proponents of the method argue that this overestimation
of direct cost compensates for the omission of indirect costs (Borestein, Beshnell, & Wolak,
2002) and (Bose, Shukla, Srivasta, & Yaron, 2006). Self-assessments based on business
surveys may be inclined to strategic misrepresentation (Pasha, Ghaus, & Malik, 1990). The
reported outage cost may be an exaggeration to impress upon the power company the
need for more reliable electricity. Alternatively, the interviewees may be unaware of the
cost or unable to devote the necessary time to complete the questionnaire.

Despite these shortcomings, the flexibility of the DAM and its link to observable market
behavior recommends its use in outage cost research (Pasha, Ghaus, & Malik, 1990). In this
study, in order to arrive at the total cost of load shedding, the DAM was used. This was due
to the nature of the study, research has shown that there is poor record keeping among small
scale enterprise (Bancy, 2007). Thus, DAM was found to be appropriate as the data required
for the study could easily be collected. The other methodologies would require the use of
more detailed data which might have proved difficult to collect given the time frame of the
study.

Table 2 shows variables used for estimating cost of load shedding to small scale
enterprises using the DAM theoretical framework.

Table 2: Variables for estimating cost of load shedding to small scaleenterprise

Independent variable Description
Material Destruction Cost • Average costs of materials lost due to load shedding.

Labour Cost • Labour cost paid due to idle labour as a result of load shedding
• Labour cost paid due to over time as a result of load shedding

Equipment damage cost • Costs associated with equipmentdamage due to load shedding.
Maintenance cost • Costs associated with maintenance of alternative sources of
energy.
Restart costs • Costs associated with re-startingoperations due to loss of power.

13
 14
Source: SMES Poporn Website, Zambia
 CHAPTER 3

METHODOLOGY

3.1 Approach and Methodology

The study employed both qualitative and quantitative survey techniques to collect data
using face to face interviews. The semi structured questionnaire used in the study was
developed by the Energy Regulation Board Zambia (ERB) in consultation with the
Regional Electricity Regulators Association (RERA) of the Southern Africa and the Central
Statistical Office (CSO) Zambia. The final questionnaire consisted of two main parts; the
first part consisted of questions on identification particulars of small scale enterprises
while the other part consisted of questions on the operations of the business. The research
questionnaire was pre-tested on a small group before it was finalised. The research assistants
underwent a training session to acquaint themselves with the questionnaire. Collection of
survey data was over a period of 3 weeks in September 2016.

3.2 Sample Survey Coverage and Target population

The survey on the impact of load shedding covered small scale enterprises found in four
cities namely; Lusaka, Kitwe, Ndola and Livingstone. The four cities were purposively
sampled owing to their relatively higher concentration of economic activities in the
sectors of interests in addition to the ease by which the cities could be accessed with the
available resources.

For the purpose of this study, an enterprise was defined as an undertaking engaged in the
manufacturing or provision of services or any undertaking carrying out business in the
4
field of manufacturing, construction and trading services . Further, based on the CSO
classification of small scale business enterprises, the study was limited to establishments
whose annual turnover did not exceed ZMW 250,000.00 in 2011.

4
[Link]

15
3.3 Sampling Design

3.3.1 Sampling frame

The 2011 SBR was used as the sampling frame for the survey. The SBR comprises a list of
business establishments in the country classified into three mutually exclusive and
exhaustive categories using respective annual turnover as a measure of size. The
categories are as follows:

i. Large-scale – consists of Business establishments whose annual turnover is ZMW

800, 001 or more.

ii. Medium-scale – consists of Business establishments whose annual turnover is

from ZMW 250, 001 to ZMW 800,000.

iii. Small-scale – consists of Business establishments whose annual turnover is ZMW

250,000 or less.

For this survey, only the small scale part of the SBR was adopted as an intact sampling frame.

3.3.2 Sample Size Determination and Allocation

In this study the sample was drawn from a population of 15,415 small scale enterprises. The
determination of the sample size was mainly guided by the need to strike a balance between
the desired sampling accuracy and its associated cost. For the purpose of this exploratory
survey and due to financial constraints, an error margin of about 9.5 percent was
set as tolerable. Typically the smaller the margin of error with it’s associated Coefficient of
Variation (CV), the larger the sample. Nonetheless, a good sampling design is not only
seen in terms of sampling accuracy but also in terms of how it brings financial and human
resource, and logistics requirements to manageable levels. In the case of this sample design,
a margin of error of 9.5 percent is associated with CVs of less than 20 percent. As a rule of
thumb, any estimate that is associated with the CV of 20 percent and below is acceptable
(Kish, 1965).

Given the above sample specifications, a sample of 696 small scale enterprises was
determined as desirable. This sample took into account a margin of error of 9.5 percent, a
design effect of 1.3 and a non-response rate of 20 percent. Ultimately, the Impact of Load
shedding survey enumerated 600 small scale establishments, representing a response rate
of 86 percent.

3.3.3 Sample weights and Sampling

Sampling weights were required to correct for differential representation of the sample
due to the disproportionate allocation of the sample and to make inference to the rest of

16
the population. The weights of the sample are equal to the inverse of the product of the
selection probabilities employed. As stated earlier, the districts were selected purposefully
and all sectors in each district were represented in the sample. Therefore, weights of the
sample in this case were equal to the inverse of the section probability of an entity within a
sector in each district.

The selection probability of an enterprise was calculated as follows:

nei
Pei = ---------------
Nei

Where

Pei = the selection probability of an entity

n = the number of enterprises selected from the ith sector in the district.
ei

Nei = Total number of enterprises listed in the district.

Therefore, the sector specific sample weight was calculated as follows: Wi = 1/ Pei

Wi is called the Probability Proportion to Size sample weight.

In order to reflect growth in the population, the base weight was adjusted using a post
stratification adjustment factor as follows:

Wf = Wi x adj f,

where adj f is obtained by dividing the projected population by the survey population.

3.4 Data analysis and techniques

The data was analysed using Statistical Package for Social Sciences (SPSS) software. Data
entry was done in CSpro which was later exported into SPSS for analysis.

3.5 Limitations

The study faced a number of limitations, firstly the sample used in the study was based on
the CSO’s SBR which was last updated in 2011. Therefore, this presented a challenge as the
sampled respondents had either closed operations or shifted their business operations
elsewhere and could not be located. This observation is in line with Mason (2009) who
observed that the average life cycle of small scale enterprises is around five years or less. In
order to overcome this challenge, sampling with replacement was employed. Additionally,
some addresses on the sampling frame were not clear to locate and even worse, the frame
didn’t have phone numbers to enable enumerators to make prior arrangements and

17
assist in locating establishments. This impacted negatively on turn-around time as the
enumerators spent unusually longer time locating establishments.

Further, the nature of the survey and timing of the study also affected the response
rate. The topic the survey was addressing was sensitive and some of the establishments
felt that some of the information, especially financial data, was confidential and hence the
enumerators faced challenges to collect the information. This factor contributed to reduced
turn-around time in data collection. In terms of timing, the survey coincided with the
national elections which in some cases provided challenges of respondent’s cooperation.

18
 CHAPTER 4

RESULTS AND DISCUSSION OF FINDINGS

4.1 Distribution of small scale enterprises by sector

In 2015, there were a total of 15,415 small scale enterprises, excluding those in mining or
recovery of minerals, of which 9,219 were based in Lusaka, 2,091 in Kitwe, 2,989 in Ndola
and 1,116 in Livingstone. The study established that 64.1% of the enterprises were
formalised by a way of registration with one of the one of the local authorities such
as Zambia Revenue Authority (ZRA), National Pension Authority (NAPSA), PACRA and the
Local Council. In addition, 27% of the enterprises have social security schemes or do make
contributions on behalf of their employees to NAPSA, LASF, PSPF and any other social
security system. This means that any disruption in business operations that affects labour
contributions to social security systems would impact one third of the enterprises.
Business formalisation is more likely to take place in urban areas mainly involving large
firms and those already using proper book keeping (Coolidge & Ilic, 2009). Registration
of the business with authorities does indicate evidence of record keeping.

Figure 2 below shows the distribution of small scale enterprises in Lusaka, Livingstone,
Kitwe and Ndola in 2015.

Figure 2: Number of small scale enterprises by location - 2015

10,000  
9,000  
8,000  
7,000  
6,000  
5,000  
4,000  
3,000  
2,000  
1,000  
-­‐  
Kitwe Ndola Lusaka Livingstone
Number  of  Enterprises 2,091 2,989 9,219 1,116

Among the enterprises that were interviewed the period of establishment ranged from 1922
to 2016. The enterprises that were established after 1991, when the economy was liberalized
accounted for 90.9% of the enterprises. The study therefore captured enterprises that

19
was liberalized accounted for 90.9% of the enterprises. The study therefore captured
enterprises that were mature and could be assumed to be conversant with their
operations. Among the enterprises, the majority were not insured for instance only in
2014
were and 2015
mature 17.8%
and couldand 18.2% were
be assumed the
to be only oneswith
conversant insured
theirrespectively.
operations. For thosethe
Among
who reportedthe
enterprises, being insured,
majority thenot
were common
insuredinsurance wasonly
for instance property equipment
in 2014 which
and 2015 was
17.8% and
18.2% were
reported the
by at only ones
48.4%, insured
followed respectively.
by motor vehiclesFor
at those who
46.7%, reported
followed by being insured,
followed by lifethe
common
at insurance
3.9 % and was
other at property equipment which was reported by at 48.4%, followed by
0.9%.
motor vehicles at 46.7%, followed by followed by life at 3.9 % and other at 0.9%.

The
The enterprises
enterprises were
were requested
requestedto
torank
rankthe
themajor
majoroperational
operationalconstraints
constraintsand
andstated
statedas as
depictedininfigure
depicted figure3.
3.

Figure 3:
Figure 3: Enterprises
Enterprises operational
operational constraints
constraints --2015
2015
40.00

35.00

30.00

25.00
Percentage

20.00

15.00

10.00

5.00

-
Competiti
Finance Fuel Electricity Labour Security Other
on
Percentage 27.70 1.00 35.80 4.00 20.40 3.50 7.60

The highest ranking operational constraint was electricity which was reported by 35.8%
The highest ranking operational constraint was electricity which was reported by 35.8%
of the enterprises in the survey. This was followed by Finance at 27.7%, competition,
of the enterprises
security in the were
and other which survey. This was
reported followed
by 20.4%, 3.5%byand
Finance at 27.7%, competition,
7.6% respectively. In a study on
security andinsecurity
Electricity other which
andwere
SMEs,reported by 20.4%,
the Overseas 3.5% and 7.6%
Development respectively.
Institute In a
(UK) established
that 49.3%
study of SMEs in
on Electricity Sub Saharan
insecurity and Africa,
SMEs, identified electricity
the Overseas has a major
Development constraint
Institute (UK) in
their business operations (The Overseas Development Institute, 2014).
23
In terms of the legal status of the establishment, the majority (62%) were individual
proprietorship followed by private limited companies (23.1%) and partnerships (7.2%). The
rest accorded for 7.7%.

Table 3 shows the distribution of small scale enterprises in Lusaka, Kitwe, Ndola and
Livingstone in 2015 by location and economic sector. Wholesale and retail trade had the
highest number of small scale enterprises at 7,533 (48.9%), followed by accommodation and
food services and other services activities at 2,771 (18.0%) and 1,868 (12.1%) respectively.

20
Table 3: Distribution of small scale enterprises by sector – 2015
Percent
Sector / District Kitwe Ndola Lusaka Livingstone Total age
Share
Agriculture, Forestry and - 1 - 11 12 0.1%
Fishing
Mining Support Services 32 - - - 32 0.2%

Manufacturing 136 195 1,023 72 1,426 9.3%

Repair and Installation of 15 - 56 - 71 0.5%

Machinery and Equipment
Water Supply Sewerage 0.1%
Waste Management and - - 11 - 11
Remediation
Construction 28 - 38 - 66 0.4%

Wholesale and Retail Trade 1,088 1,445 4,440 560 7,533 48.9%

Transportation and Storage 26 - 86 25 137 0.9%

Accommodation and Food 396 696 1,479 200 2,771 18.0%

Services
Information and 30 26 135 - 191 1.2%
Communication
Real Estate Activities - 9 41 - 50 0.3%

Professional Scientific and - 30 164 - 194 1.3%

Technical Activities
Administrative and Sup- 49 72 176 23 320 2.1%
port service activities
Education 90 90 342 46 568 3.7%

Human health and So- - - 84 14 98 0.6%

cial work activities
Arts Entertainment 6 9 52 - 67 0.4%
and recreation
Other services activities 195 416 1,092 165 1,868 12.1%

Total 2,091 2,989 9,219 1,116 15,415 100.0%

4.2 Number of workers employed by small scale enterprises

Table 4 shows the number of workers employed by small scale enterprises by gender and
location. Small scale enterprises provided employment to a total of 174,028 employees in
2015. This implies that each establishment on average had 11 employees. In terms of sex,
70,644 (41%) were female and 103,383 (59%) were male. In terms of location, Lusaka district
had the highest number of employees at 113,385 (65%), followed by Kitwe 26,530
(15%) and Ndola 24,101 (14%) in that order. Livingstone had the least number of employees
at 10,011 employees reflecting 6%.

21
 Table 4: Number of workers employed by small scale enterprises
Female employees 2015 Male employees 2015 Total employees 2015
Kitwe 6,794 19,736 26,530
Ndola 9,756 14,344 24,101
Lusaka 51,165 62,220 113,385
Livingstone 2,929 7,082 10,011
Total 70,644 103,383 174,028

4.3 Wage bill

The annual wage bill for the establishments ranged from K136 to K466,869.45 in 2015.
The average wage bill for each enterprise was K13,474.33 per month. The wage bill
broken down by city for 2015 and 2014 is depicted in figure 4.

Figure 4: Small scale enterprises annual wage bill – 2014 and 2015
1,800,000,000  
1,600,000,000  
1,400,000,000  
1,200,000,000  
1,000,000,000  
ZMW  

800,000,000  
600,000,000  
400,000,000  
200,000,000  
0  
Kiwe Ndola Lusaka Livingstone Total
Annual  Wagebill  2014 305,471,077 262,556,339 813,980,069 116,976,256 1,498,983,742
Annual  Wagebill  2015 281,456,809 274,918,877 920,915,623 97,436,291 1,574,727,600

The annual wage bill increased from K1,498,983,742 in 2014 to K 1,574,727,600 in

2015 reflecting an increase of 5.1%. In terms of average monthly earnings, an employee
earned K754 in 2015. This figure was lower than the average earning of K 2,344 per
month which was captured by the CSO’s labour force survey for 2014 for both formal
and informal sector (Central Statistical Office, 2015).

4.4 Business working hours

Table 5 shows the average number of working hours per day for small scale enterprises
by location in 2015. Lusaka, Ndola and Livingstone districts had the highest number of
working hours per day of 11 hours, with Kitwe having the least number of working

22
hours of 10 hours. Among the enterprises, only 3.7% operated their businesses beyond
22 hours implying that for any load shedding that was done beyond 22 hours would
have had an impact of their operations.

Table 5: Average number of operating hours per day

District Average operating hours per day 2015
Kitwe 10
Ndola 11
Lusaka 11
Livingstone 11
Total 11

Table 6 shows the average number of operating days per week for the small scale
enterprises in the four districts. On average, the small scale establishments operated 6 days
in a week as summarized in table 5.

Table 6: Average number of operating days per week - 2015

District Average number of operating days per week

Kitwe 6
Ndola 6

Lusaka 6

Livingstone 6

Total 6

4.5 Load shedding Experience

Table 7 shows the number of small scale enterprises that had experienced load shedding in
2015. The table shows that 12,781 (83.1%) of the small scale enterprises experienced load
shedding in 2015, of these, the highest number at 8,343 (65%) were from Lusaka followed
by Ndola with 1,874 (13%) and Kitwe with 1,713 (12%). The lowest number was Livingstone
with 851 (7%). Meanwhile, those who reported not experiencing load shedding 59.0% did
not use electricity for their business operations.

23
Table 7: Number of small scale enterprises that had experienced load shedding - 2015
Load shedding experience during the year Total
2015
yes no
District Kitwe 1,713 378 2,091
Ndola 1,874 1,115 2,989
Lusaka 8,343 843 9,186
Livingstone 851 265 1,116
Total 12,781 2,601 15,382

In terms of the average number of hours of load shedding experienced per day, overall the
total number of hours of load shedding per day increased from 1 hour per day in 2014
to 4 hours per day in 2015, representing a percentage increase of 300% as summarized in
Table 8. In terms of the average, enterprises experienced load shedding hours of 3.9 hours
from January to June and 6.9 hours from July to December in 2015. Therefore, most load
shedding hours were experienced during the period July to December 2015. Around this
time it was reported that the power deficit in 2015 ranged from 560 to 1000 MW.

In 2015, the areas that experienced the most load shedding hours on average per day
was Lusaka at 5 hours followed by Livingstone at 4 hours. Kitwe and Ndola on average
experienced 3 hours per day. This is summarized in table 8.

Table 8: Average number of hours of load shedding per day – 2014 and 2015
Average Load shedding hours Average Load shedding hours per day
District per day 2015 2014
Kitwe 3 1
Ndola 3 1
Lusaka 5 1
Livingstone 4 0
Total 4 1

As depicted in table 8 the extent of load shedding, in the four cities intensified in 2015
compared to 2014.

During 2015, load shedding schedules were developed by the Utility advising the public
on the time and day when the power outage would be implemented. Enterprises were
requested to state if this schedules were strictly followed by ZESCO and 51% stated that
they were followed while 49% said otherwise.

Table 9 shows the source of information for load shedding schedules by location in
2015. The table shows that most (5,746) small scale enterprises indicated that the
Short Messaging System (SMS) was the major source of information reflecting 45.0%
followed by members of the public at 2,425 (19%). Similarly, a total of 724 (6%)

24
indicated that they received this information from other sources which included ZESCO
sales points, through own observation and leaflets. A total of 2,125 (17%) indicated that
they did not receive information on ZESCO load shedding schedules.

Table 9: Small scale enterprise source of information for ZESCO’s load shedding schedule
Major source of ZESCO load shedding schedules
Members
of the
public
District SMS Radio TV Newspaper Other None Total
Kitwe 560 51 67 167 749 85 34 1,713
Ndola 705 34 60 210 315 32 518 1,874
Lusaka 4,379 221 45 810 1,049 266 1,573 8,343
Livingstone 102 0 96 0 312 341 0 851
Total 5,746 306 268 1,187 2,425 724 2,125 12,781

4.6 Electricity expenses by small scale enterprises

In this study, electricity expenses ranged from K50 to K 14,083 per month. The average
electricity expenditure was K754.61. Past studies on Zambia which captured the electricity
expenses by small scale enterprises were not immediately available. In this study, the
highest expenditure was recorded in October at K10,427,260 while the least was recorded
in December at K9,102,827. Table 10 shows the total annual expenditure on electricity
by small scale enterprises in 2014 and 2015. The table shows that expenditure on
electricity increased by 3.0% from K 114, 897,354 in 2014 to K 118,307,145 in 2015.

Table 10: Small scale enterprises Expenditure on electricity – 2014 and 2015
District Electricity Expenses in 2014 Electricity Expenses in 2015
Kitwe 12,228,442.00 12,099,950.00
Ndola 34,036,218.00 37,279,467.00
Lusaka 62,702,574.00 62,469,328.00
Livingstone 5,930,120.00 6,458,400.00
Total 114,897,354.00 118,307,145.00

In terms of the proportion of electricity expenses to the total annual business expenses
in 2015, the majority (85%) of the small scale enterprises on average indicated that
electricity bills constituted up to 25 % of their total annual business expenses. This was
followed by 10% who indicated that electricity bills constituted between 26 to 50 %
proportion of the total business expenses, while 5% indicated above 51%. This information
is summarized in Table 10.

25
 Table 11: Cost of electricity as a proportion of total annual expenses in 2015
Cost of electricity in Kwacha as a proportion of the total
expenses in 2015

0 -25 percent 26-50 percent 51-75 percent 76-100 percent Total

District Kitwe 1799 157 0 0 1,956

Ndola 2044 258 41 0 2,343
Lusaka 6846 906 493 90 8,335
Livingstone 811 0 0 0 811
Total 11,500 1,321 534 90 13,445

4.7 Business Annual Turnover

In this study, the annual turnover, defined as total sales of the business in a year,
ranged from K 2,200 to K10,999,989. The average annual turnover was K407,527, while the
highest average turnover per month was recorded in April at K41, 205.68, while the least
turnover was recorded in November K33, 849.50. In 2015, in terms of breakdown by cities
the highest total turnover (K3,068,940,237) was recorded in Lusaka followed by Kitwe at
K1204,479,503, Ndola at K853,639,713 and Livingstone at K650,861,[Link] performance
of turnover in 2015 against 2014 is depicted in table12. Generally in all the cities, there
was a decline in nominal turnover, overall it declined by 7.1%. Similarly, in real terms, the
annual turnover declined from K6, 222,368, 193 in 2014 to K4, 921,568,488 in 2015 reflecting
a reduction of 20.9%.

Table 12: Small scale enterprises turnover – 2014 and 2015

District 2015 (Real) 2015 (Nominal) 2014

Kitwe 1,025,962,098.00 1,204,479,503.00 1,443,210,307.00

Ndola 727,120,710.00 853,639,713.00 929,993,511.00

Lusaka 2,614,088,819.00 3,068,940,273.00 3,169,717,225.00

Livingstone 554,396,861.00 650,861,915.00 679,447,150.00

Total 4,921,568,488.00 5,777,921,404.00 6,222,368,193.00

The use of electricity by enterprises is depicted in Table 13. In 2015 enterprises that used
electricity in their business accounted for 88.3% (13,542) compared to 85.3% (13,047)
in 2014. The highest percentage was recorded in Lusaka city (62.3%) followed by Ndola at
17.4%. Kitwe and Livingstone reported 13.9% and 6.3% respectively. Comparing 2014
and 2015, generally there was a marginal increase in the use of electricity by small scale
enterprises in their business operations.

26
Table 13: Electricity use by establishments for business operations
Did your establishment use electricity in itsoperations?

2014 2015
Yes No Yes No
Kitwe 1,870 203 1,888 203
Ndola 2,308 681 2,360 629

District Lusaka 8,039 1,073 8,443 698

Livingstone 830 286 851 265
Total 13,047 2,243 13,542 1,795

In terms of internet connectivity, Table 14 depicts the number of small scale enterprises
that had internet connectivity. Out of the 15,340 enterprises only 4,298 (28%) had internet
connectivity.

Table 14: Internet connectivity by small scale enterprises - 2015

Does this establishment have internet connectivity?
yes No Total
District Kitwe 811 1,262 2,073
Ndola 740 2,249 2,989
Lusaka 2,397 6,765 9,162
Livingstone 350 766 1,116

Total 4,298 11,042 15,340

4.8 The impact of load shedding on turnover

In order to determine the impact of the load shedding on the business turnover, small scale
enterprises were asked to quantify the loss in turnover in monetary terms. Load shedding
leads to the disruption of business operations that results in loss of turnover. Table 15 shows
the reported loss in turnover as a result of load shedding ranged from K0 to K759,000. On
average the reported loss in turnover was K19,251.16 as depicted in table 15. The total
reported annual loss in turnover from the four cities in 2015 was K456,255,094.50 reflecting
7.33% of the total annual turnover. Lusaka district recorded the highest loss in turnover of
K 296,035,506 followed by Kitwe at K86,972,776 and Ndola at K 63,718,087.50. Livingstone
had the lowest loss in turnover at K 9,528,725.

27
Table 15: Impact of load shedding on turnover - 2015
Reported Loss in turnover 2014 (Nominal) Percentage of reported
District
as a result of load losses to Turnover
shedding
Kitwe 86,972,776 1,443,210,307.00 6.03%

Ndola 63,718,087.50 929,993,511.00 6.85%

Lusaka 296,035,506 3,169,717,225.00 9.34%

Livingstone 9,528,725 679,447,150.00 1.40%

Total 456,255,094.50 6,222,368,193.00 7.33%

4.9 The Impact of load shedding on labour costs

In some cases, due to load shedding, some establishments experienced idle labour
while some incurred overtime labour costs as they were forced to operate longer hours
than usual. In the survey, 22.8% of establishments reported cases of idle labour while 9.8%
reported incurring overtime labour costs due to load shedding. .

The costs arising from idle labour were defined as the amount of money paid to a
worker otherwise not working due to load shedding. Meanwhile, overtime labour cost
was defined as the amount of money paid to a worker for extra hours worked after
load shedding. Table 16 shows the costs associated with idle labour and overtime in each
of the four districts. At firm level idle labour costs on average ranged from K0.00
to K8,333.33 per month. The average idle labour cost per firm was K 130.80 per month.
Equally, at firm level overtime labour costs on average ranged from K0 to K5, 000 per month.
The average overtime labour cost per firm was K30.49 per month.

Table 16: Labour costs due to load shedding by district

District Idle labor cost (ZMW) Overtime cost (ZMW) Total

Kitwe 3,773,150 1,427,900 5,201,050

Ndola 4,753,629 408,000 5,161,629

Source: SMES Welders Website, Zambia

Lusaka 13,138,908 3,805,450 16,944,358

Livingstone 2,529,000 0 2,529,000

Total 24,194,687 5,641,350 29,836,037

28
 29
Source: SMES Welders Website, Zambia
Lusaka experienced the highest costs of idle labour and overtime costs at K13,138,908 and
K3,805,450 respectively. Ndola recorded the second highest costs in terms of idle labour
cost of K4,753,629, while on the other hand, its overtime costs were the second lowest
after Livingstone at K408,000. Kitwe reported idle labour costs of K3,773,150 and over
time costs of K1,427,900. Livingstone recorded K2,529,000 idle labour cost and did not
experience any overtime costs. Notably, some small scale enterprises adapted to the
load shedding schedules and staff was meant to operate during the period when there
was power. In addition, some small scale enterprises employed other alternative sources of
energy.

4.10 Equipment damage and maintenance attributed to load shedding

Among the enterprises surveyed, 29.9% reported damaged equipment due to load
shedding. Furthermore, only 11.7% reported that such equipment was insured. The cost of
damaged equipment ranged from K2 to K56,000. The average cost of damaged equipment
was K 3,112.80.

A number of small scale enterprises lost equipment due to load shedding. Some of the
equipment was completely damaged. Table 17 shows costs associated with equipment
damage and maintenance costs of as a result of load shedding in 2015.

Table 17: Cost of equipment damage and maintenance costs due to loadshedding
District Cost of equipment Maintenance costs Total

Kitwe 3,394,610.00 5,629,350.00 9,023,960.00

Ndola 3,111,700.00 10,171,582.00 13,283,282.00

Lusaka 9,666,138.00 91,740,149.00 101,406,287.00

Livingstone 839,000.00 4,986,250.00 5,825,250.00

Total 17,011,448.00 112,527,331.00 129,538,779.00

A total of K17,011,448 was spent on equipment damage as a result of load shedding by small
scale enterprises. Of this amount, Lusaka city recorded the highest total cost of K9,666,138
followed by Kitwe city at K3,394,610, while Ndola city reported K3,111,700 and Livingstone
with K839,000. In terms of maintenance costs, a total of K112,527,331 was spent in the four
cities. Lusaka city recorded the highest at K91,740,149 followed by Ndola at K10,171,582,
while Kitwe reported K5,629,350 and Livingstone with K4,986,250.

In order to mitigate the impact of load shedding on the business, some small scale
enterprise opted to purchase or hire alternative energy sources such as genset, solar
panels, and invertors among others. Table 18 shows the summary of the costs associated with
purchasing and hiring per month.

30
 Table 18: Purchases and hiring costs for alternative energy sources
District Purchase Costs Hiring Costs Total
Kitwe 3,308,900 408,000 3,716,900
Ndola 6,419,672 171,300 6,590,972
Lusaka 70,991,261 1,881,360 72,872,621
Livingstone 3,630,550 588,000 4,218,550
Total 84,350,383 3,048,660 87,399,043

A total of K87,399,043 was spent on purchasing and hiring alternative energy sources by small
scale enterprises in the four districts. Of this amount, K84, 350,383 was spent on purchasing
of alternative energy sources, while K3,048,660 was spent on hiring.

Lusaka district reported the highest total cost at K72,872,621 followed by Ndola at
K6,590,972 and Livingstone at K4,218,550. Kitwe had the lowest cost at K3,716,900.

4.11 The cost of restarting operations as a result of load shedding

Business owners are likely to suffer from costs incurred due to the restarting of an
industrial process. For instance, some operations like bakeries incur costs when operations
are restarted. Further, internet café business might also incur restart costs as work might be
lost due to loss of power. The costs of restarting operations ranged from K0 to K 35,000. On
average the reported costs due to restarting of operations was K 4,663.84. Figure 4 shows
the costs due to the restarting of business operations by small scale enterprises.

Figure 5 : Costs of restarting operations due to load shedding

4,000,000.00

3,500,000.00

3,000,000.00

2,500,000.00
ZMW

2,000,000.00

1,500,000.00

1,000,000.00

500,000.00

-
Kitwe Ndola Lusaka Livingstone Total
restart cost 1,288,960.00 236,300.00 2,139,580.00 89,550.00 3,754,390.00

31
A total of K3,754,390.00 was spent on restarting of operations by small scale enterprises
in 2015 in the four cities as a result of load shedding. Lusaka recorded the highest cost of
restarting operations at K2,139,580 (57%) followed by Kitwe at K1,288,960 (34.3%%). Ndola
and Livingstone recorded K236,300 (6.3%) and K839,000 (2.4%) respectively.

4.12 Measures put in place by enterprises to tackle reduced load.

In order to cope with the increased load shedding experienced in 2015, small scale
enterprises adopted several strategies with the objective of mitigating against the impact
of load shedding on their businesses. One of the objectives of the study was to establish
the different coping mechanisms used by small scale enterprises to mitigate against load
shedding. This section outlines the various coping mechanisms employed by small scale
enterprises.

4.12.1 Demand side management strategies

In the study, 55% reported employing strategies to mitigate against load shedding.
However, there was no reported shutdown of operations and very few reallocations of
businesses (0.2%). Table 16 depicts the number of small scale enterprises that used
various strategies to cope with the reduced load by city.
Source: Energy savers Inc. Website, Zambia

32
Table 19: Demand side management strategies
Did you employ demand side management strat-
egies during load shedding experienced in 2015
such as use of LED bulbs?
Total
Yes No
District Kitwe 1,161 552 1,713
Ndola 879 978 1,857
Lusaka 4,407 3,784 8,191
Livingstone 485 366 851
Total 6,932 5,680 12,612

4.12.2 Use of generators

Figure 5 shows the number of small scale enterprises that used generators as an
alternative source of energy during load shedding in 2015. A total of 3,511 (29.9%) of small
scale enterprises used generators as an alternative source of energy, while 8,251 (70.1%)
did not. In a similar study, Andrew, Emily, Alberto and Juan (2014) established that due to
electricity insecurity, 33% of SMEs in developing countries used a generator as the main
practice to mitigate the impact of electricity insecurity.

Figure 6: Number of small scale enterprises using generators - 2015

9,000  
8,000  
7,000  
6,000  
Number  

5,000  
4,000  
3,000  
2,000  
1,000  
-­‐  
Kitwe Ndola Lusaka Livingstone Total
Yes 488 542 2,089 392 3,511
No 1,157 1,100 5,570 424 8,251

4.12.3 Use of uninterruptible power supply

Table 20 shows that out of 11, 698 respondents, only 276 (2.4%) indicated that they used
uninterruptible power supply (UPS) as an alternative energy source to mitigate against the
impact of load shedding.

33
 Table 20 : Number of small scale enterprises using uninterruptible power supply - 2015
Use UPS
Yes No Total
District Kitwe 68 1,577 1,645
Ndola 0 1,642 1,642
Lusaka 208 7,422 7,630
Livingstone 0 781 781
Total 276 11,422 11,698

4.12.4 Surge protectors

Table 21 shows the number of small scale enterprises using surge protectors as a measure
to protect electrical devices from damage among the small scale enterprises in 2015. The Table
shows that only 249 (2.1%) indicated that they used surge protectors in the four districts.

Table 21: Number of small scale enterprises using surge protectors

Yes Surge protectors
No Total
District Kitwe 68 1,577 1,645
Ndola 17 1,625 1,642
Lusaka 164 7,466 7,630
Livingstone 0 781 781
Total 249 11,449 11,698

4.12.5 Back up data systems

Table 22 shows the use of back up data systems among small scale enterprises in 2015.
In order to save data that could result from damaged electrical equipment due to load
shedding some small scale enterprises invested in back up data systems. Table 18
shows that 813 (6.9%) of the small scale enterprises invested in back up data systems.

Table 22: Number of small scale enterprises using back up data systems
Back up data systems
Yes No Total
District Kitwe 18 1,627 1,645
Ndola 290 1,352 1,642
Lusaka 505 7,125 7,630
Livingstone 0 781 781
Total 813 10,885 11,698

34
4.12.6 Use of back up batteries

Table 23 shows the use of back up batteries as a mitigation measure against load
shedding among small scale enterprises in the four districts. The table shows that out of
a total of 11,698 small scale enterprises, 803 (6.9%) indicated that they used back up
batteries in order to mitigate against the loss of power.

Table 23 : Number of small scale enterprises using back up batteries

Back up batteries
Yes No Total
District Kitwe 34 1,611 1,645
Ndola 273 1,369 1,642
Lusaka 496 7,134 7,630
Livingstone 0 781 781
Total 803 10,895 11,698

4.12.7 Security enhancement

Load shedding brings about security concerns for small scale enterprises. In order to cope
with theft threats, small scale enterprises enhanced their security features of their business.
Table 24 shows that 803 (6.9%) small enterprises reported that they had enhanced their
security features in the four districts.

Table 24: Security enhancement

Security enhancement
Yes No Total
Dis- Kitwe 51 1,594 1,645
trict
Ndola 389 1,253 1,642
Lusaka 349 7,281 7,630
Livingstone 14 767 781
Total 803 10,895 11,698

4.12.8 Reduction in labour

Small scale enterprises indicated that one way of coping with load shedding was through
the reduction in labour. Table 25 shows that 1,013 (8%) out of 12,452 small scale enterprises
indicated that they reduced the labour hours compared to 11,439 (92%) who did not.

35
Table 25 : Number of small scale enterprise that reduced labour hours due to load shedding
- 2015
Did the establishment reduce on labour to either
cope or remedy the load shedding situation?
Total
Yes No
District Kitwe 250 1,463 1,713
Ndola 136 1,704 1,840
Lusaka 574 7,474 8,048
Livingstone 53 798 851
Total 1,013 11,439 12,452

As stated earlier, each establishment on average had 11 employees, however, during

the period of load shedding several of them laid off their work force. Specifically, it was
reported that on average 15 workers were laid off by each establishment at an associated
average cost of K14,883.42. This cost relates to redundancy costs.

4.12.9 Reduction in the working hours

Table 26 shows the number of small scale enterprises that had reduced their working hours
due to load shedding. A total of 876 (7.6%) indicated that they reduced business working
hours due load shedding.

Table 26: number of small scale enterprises that reduced working hours due to load
shedding - 2015
Did the establishment reduce on working hours
to either cope or remedy the load shedding
Total
situation?
Yes No
District Kitwe 73 1,640 1,713
Ndola 155 1,685 1,840
Lusaka 593 7,426 8,019
Livingstone 55 796 851
Total 876 11,547 12,423

4.12.10 Relocation of business

Table 27 shows the number of small scale enterprises that relocated their business due to
load shedding. The table shows that only 21 (0.2%) small scale enterprises out of 12,402
that had responded indicated that they relocated their business to a different location
due to load shedding. These enterprises were all from Lusaka.

36
 Table 27 : Number of small scale enterprises that relocated due to load shedding - 2015
Did the establishment Relocate to either
cope or remedy the load shedding situation? Total

Yes No
District Kitwe 0 1,713 1,713
Ndola 0 1,840 1,840
Lusaka 21 7,998 8,019
Livingstone 0 851 851
Total 21 12,402 12,423

In terms of costs associated with relocation of the businesses, it was reported thata total of
K 390,191 was spent on relocation of business reflecting an average cost of K 18,580.52 per
each establishment.

4.12.11 Shutdown of business operations

Table 28 shows that none of the small scale enterprise had reported shutting down their
business operations due to load shedding.

Table 28: Number of small scale enterprise that had shutdown business operations due to
load shedding - 2015
Did the establishment Shutdown to either
cope or remedy the load shedding situation?
Total
No
District Kitwe 1,713 1,713
Ndola 1,840 1,840
Lusaka 8,019 8,019
Livingstone 851 851
Total 12,423 12,423

4.12.12 Change of operations hours

ZESCO designed load shedding schedules in order to assist its clients to plan. The
respondents were asked to indicate if they switched the operating hours in order to suit
the load shedding schedule. Table 29 shows that 808 (6.5%) of the small scale enterprises
from three districts indicated that they changed their business operating hours in order
to accommodate the load shedding schedule. However, none of the small scale enterprises
in Livingstone had indicated change of operating hours.

37
 Table 29 : Switching of working hours
Did the establishment switch working hours as a
measure to either cope or remedy the load
shedding situation?
Total
Yes No
Dis- Kitwe 152 1,561 1,713
trict
Ndola 182 1,658 1,840
Lusaka 474 7,545 8,019
Living- 0 851 851
stone
Total 808 11,615 12,423

A total of K 1,956,735 was spent on the switching of working hours as a measure to cope
with load shedding reflecting an average cost of K2,421.70 per establishment.

4.12.13 Other measures taken

Table 30 shows other measures taken by small scale enterprises to mitigate against the
impact of load shedding which included the use of power banks, candles, charcoal, use of
solar and rechargeable lamps, use of torch and additional casual labour for security and
doing laundry in some instances.

Table 30: Number of small scale enterprises that used other measures
Other measures taken to mitigate against
the impact of load shedding

Yes No Total
District Kitwe 405 1240 1645
Ndola 640 1002 1642
Lusaka 2236 5372 7608
Livingstone 290 491 781
Total 3,571 8,105 11,676

With regards to costs, a total of K4, 178,750 was spent on other measures translating into
average costs of K1, 170.19.

Furthermore among those who reported being load shedded, a total of 3,814 (30.1%)
reported having lost material. The details on materials lost include the following: beers,
foodstuffs, paint, among. Figure 7 shows the number of small scale enterprises that indicated
loss of materials in 2015 per district.

38
   Figure 7: Number of enterprises that lost materials due to load shedding
 
10,000  
 
  9,000  
  8,000  
  7,000  
  6,000  
 
  5,000  
  4,000  
  3,000  
  2,000  
 
1,000  
  -­‐  
Kitwe Ndola Lusaka Livingstone Total
Yes 699 595 2,321 199 3,814
No 1,014 1,262 5,937 652 8,865
 

The total cost of material lost ranged from K0 to K70,000. The average cost of material lost
was K304 per firm. Cumulatively all enterprises reported a total cost of material lost as K
4,487,214. Of the enterprises that reported material lost, only 3.2% of the establishments
reported having insured the materials that were lost due to load shedding.

4.13 Challenges due to load shedding

During load shedding establishments faced several challenges as depicted in figure 8.

Water supply disruption (85.5%) was reportedly the major challenge followed by storage
disruption (40.6%) and theft cases (20.5%) in that order. On the other hand 19.9% reported
other operational challenges which included delayed
27   finishing of works for those in tailoring
business and reduction in business sales among others. While 18.1% reported they did
18.1% reported challenges
not experience they did not
due experience challenges
to load shedding. due8 depicts
Figure to loadtheshedding.
challengesFigure
due to8
90.0%
depicts the challenges
load shedding
80.0% due by
as reported toenterprises.
load shedding as reported by enterprises.
70.0%
Figure8 8: 60.0%
Figure : Challenges
Challenges facedfaced
due todue
load to load shedding
shedding byenterprises
by small scale small scale enterprises
Percentage

50.0%
90.0%
40.0%
80.0%
30.0%
70.0%
60.0%
Percentage

50.0%
40.0%
30.0%
20.0%
10.0%
0.0%
Phone Internet Storage
Theft Water Supply Other None
disruption Disruption Disruption
Percentage 20.5% 17.6% 17.9% 85.5% 40.6% 19.9% 18.1%

4.14 The impact of load shedding on39

small scale enterprises
 4.14 The impact of load shedding on small scale enterprises

As stated in chapter two of this study report, the impact of load shedding on small
scale enterprises will be estimated using the Direct Assessment Method (DAM).

The total direct cost relationship is restated in the formula below:

TDCi = OLi + MCi + LCi + EDCi + RCi

Where:

TDCi is the total direct cost for the ith consumer;

OLi is cost of lost output (turnover);

MDCi is the material destruction cost; LCi is labour cost;

EDCi is the equipment damage;

MCi is the maintenance cost as a result of load shedding; and RCi is restart cost.

The cost per unit of electricity (kWh) lost is the estimated as follows:

TDCi
OCi =
-----------
kWhlosi

Where:

OCi is the cost per kWh lost; and

kWhlosi are the total units of electricity (kWh) lost or unsupplied due to load shedding
computed as the product of normal energy consumption per hour multiplied by the
total number of load shedding hours in a year. Table 32 shows the variables used in the
DAM.
Source: SMES Website, Zambia

40
Table 31: Variables used in the DAM
No. Symbol Variable Cost

1. OLi Cost of lost output (turnover) 456,255,094.50

2. MDCi Material destruction costs; 4,487,214.00

3. LCi Labour costs (idle and overtime costs) 29,836,037.00

4. EDCi Equipment damage 17,011,448.00

5. MCi Maintenance cost 112,527,331.00

6. RCi Restart costs. 3,754,390.00

7. TDCi Total direct cost for the ith consumer 623,871,514.5

8. kWhlosi Total units of electricity (kWh) lost or unsupplied 65,927,681.82

due to load shedding

OCi Cost per kWh lost (ZMW) 9.46

The total cost of lost output for all enterprises was K 456,255,094.50 while the cost of lost
material was K4,487,214. Further, labour costs were K29,836,037, while a total cost
of K112,527,331 related to maintenance costs and K17,011,448.00 to equipment damage
costs. The costs of restarting operations were K3,754,390. The overall total direct costs
amounted to K623,871,514.5 and total units of 65,927,681.82 of unsupplied electricity
translating into ZMW 9.46/kWhlos (kilowatt hour lost).

The district with the highest loss was Kitwe with US$ 1.94/kWhlos, followed by Lusaka at
US$ 0.97/kWhlos. Livingstone district was third at US$ 0.53/kWhlos, while Ndola was the
lowest at US$ 0.51/kWhlos.

Table 32: The cost of load shedding on small scale enterprises

District Cost per cost per kWh Total direct cost kWh lost
kWh US$ (ZMW)

Kitwe 1.94 19.43 104,981,846.00 5,403,856.00

Ndola 0.51 5.13 82,621,268.50 16,105,575.00

Lusaka 0.97 9.71 417,605,675.00 43,009,080.30

Livingstone 0.53 5.34 18,662,725.00 3,496,189.09

Overall Cost 0.95 9.46 623,871,514.50 65,927,681.82

41
 Chapter 5
Conclusion and recommendations

5 Conclusion

The Zambian Electricity Supply Industry (ESI)’s overreliance on hydro generation suffered
a setback in 2015 when there was a drastic reduction in electricity supply which was
attributed to the reduced generation by ZESCO Limited (ZESCO) due to the low water levels
in the reserves caused by poor rainfall in the 2014/15 rainy season.

The resultant power deficit of between 560 to 1000 MW resulted in load shedding of up to 8
hours. The load shedding affected the most business operations and financial viability.
The small enterprises were the most affected mainly due to their lack of resilience
and limited capacity to invest in alternative energy sources. Given the importance of small
enterprises in the economy, it is imperative that the impact of load shedding is studied and
understood so that corrective steps are undertaken.

The general demographic and characteristics of the sample established that most small
enterprises are not yet formalized although they have existed for over 10 years. The small
enterprises faced serious electricity constraints. Most enterprise provide employment to
an average of 11 employees who are given wages. Overall the total number of hours
of load shedding per day increased from 1 hour per day in 2014 to 4 hours per day in
2015, representing a percentage increase of 300%. Most enterprises were uncertain about
the loading schedules because these were not strictly followed by ZESCO. The majority
(85%) of the small scale enterprises on average indicated that electricity bills constituted
between 0 to 25 % of their total annual business expenses.

Meanwhile, between 2014 and 2015, the reported loss in turnover as a result of load
shedding ranged from K0 to K759,000. On average the reported loss in turnover was
K19,251.16. Close to a third of the establishments, experienced cases of idle labour while
and overtime labour costs due to load shedding.

Further, some enterprises reported damaged equipment due to load shedding and
most of this equipment was not insured. Meanwhile, some enterprises spend resources on
restarting of operations as a result of load shedding.

During load shedding some enterprises employed strategies to mitigate against load
shedding. However, there was no reported shutdown of operations and very few
reallocations of businesses. The key mitigating strategies were: use of generators; use of
uninterruptible power supply (UPS); use of surge protectors; use of back up data systems;
and use of back up batteries. Further, as a consequence of load shedding, some small

42
enterprises enhanced their security features while others reduced labour hours and others
changed their business operating hours.

The study established that a total cost of K623,871,514.50 was incurred as a result of load
shedding by small scale enterprises translating into US$ 0.95/kWhlos (kilowatt hour lost).
In terms of kilowatt hours lost per district, Kitwe had the highest loss at US$ 1.94/kWhlos,
followed by Lusaka at US$ 0.97/kWhlos. Livingstone district was third at US$ 0.53/kWhlos,
while Ndola was the lowest at US$ 0.51/kWhlos.

This study has established that the incidence of load shedding in 2015 lead to adverse
disruptions in the operations of most small enterprises in the survey areas. Furthermore,
most small enterprises had inadequate response strategies as they could not use alternative
sources of energy. Most small enterprises resorted to reducing their work outputs resulting
in reduced turnover whilst incurring additional costs such as idle labor and overtime. Some
enterprises suffered losses due to equipment damage and high replacement costs.
The study estimate of US$ 0.95/kWh for each unsupplied electricity unit confirms the
proposition that small enterprises were adversely affected by load shedding and that there
is a possible inverse relationship between load shedding and small enterprise productivity
as well as general business performance.
Source: SMES Website, Zambia

43
 44
Source: zambiatourism Website, Zambia Source: zambiatourism Website, Zambia
Recommendations
1. The over dependence on hydro power coupled by the inability by small enterprises
to response to any disaster in the electricity system, makes it inevitable that the
country diversifies the energy mix. The country must develop an integrated Resource
Management Plan (IRP) that will clearly outline pipeline projects for alternative
sources of energy such as solar, coal, geo thermal and wind, which are less prone
to disasters such as reduced water levels.

2. Most small enterprises are not formalized and have limited insurance which is a
risk to their business operations and financial viability. It is recommended that
Government puts in place initiatives that promote business formalization and
insurance schemes. Once formalized, it will be easy for small enterprises to
access support such as credit to access alternative sources of energy and insurance
schemes.

3. The lack of cost reflective tariffs has discouraged investment in both additional
hydroelectricity capacity and alternative sources of electricity. The cost of service
study being undertaken by the Energy Regulation Board (ERB) in 2017, must be
utilized and its recommendations implemented in full so that a clear cost of
power production is known and the migration path to cost reflectivity is clearly
understood and known by all concerned.

4. Whilst the 2015 power situation was induced by a natural phenomenon, there is
still concern that ZESCO must improve on the quality of service delivery. A case in
point is when small enterprises could not plan their business operations because
the load shedding schedules that were published by ZESCO were not strictly
followed. Additionally, 17% of the enterprises reported that they did not have
access to information on the load shedding schedules developed by the Utility.
In this regard, ZESCO must increase awareness of information on the load shedding
schedules. Further, the ERB must enhance the Key Performance

5. Indicator Framework (KPI) for ZESCO to cover the aspect of adherence to load
shedding schedules.

6. In addition, specific studies must be commissioned to ascertain the comparative

cost benefit analysis for hydroelectricity in comparison to alternative sources of
electricity.

45
 References

Andrew , S., Emily , D., Alberto , L., & Juan, P. (2014). How does electricity insecurity affect
businesses in low and middle income countries? Retrieved April 15, 2016, from [Link].
org
Bancy, M. W. (2007). Record Keeping and Growth of Micro and Small Enterprises. Retrieved
December 16, 2016, from [Link] [Link]/bitstream/handle/123456789/7122/
Bancy%20Wawira%[Link]?sequen ce=1
Borestein, S., Beshnell, J. B., & Wolak, F. A. (2002). Measuring market inefficiencies in
California’s restructured wholesale electricity market. . American Economic
Review,volume , 92, no.5, pp. 1376–1405. .

Bose, R. K., Shukla, M., Srivasta, L., & Yaron, G. (2006). Cost of unsaved energy in Karnataka,
India.

Karnataka: Energy policy Rview.

Central Intelligence Agency. (2017). The World Fact Book. Retrieved April 2017, from https://
[Link]/library/publications/the-world-factbook/geos/[Link]
Central Statistical Office. (2015). Zambia Labour Force Survey Report 2014. Lusaka: Central
Statistical Office.
Chisala, C. (2008). Institute of Developing Economies. Retrieved May 15, 2016, from http://
[Link]/English/Publish/Download/Dp/pdf/[Link]
Coolidge, J., & Ilic, D. (2009). Tax Compliance Perception and Formalisation of Small
Businesses in South Africa. Policy Reserach Working Paper 4992 World Bank, 1.
Institute of Public Policy. (2013). Introduction and Analysis of Secondary Data. Lahore:
Institute of Public Policy.
Kanlisi , S. K., Amenga, J. E., Akomeah , D. O., Amoako , R., & Narh , E. (2014). The Socio-
Economic Contribution Of Small-Scale Industries to Livelihood of Women in the Shea
Butter Industry in the Wa Municipality. European Scientific Journal, 50-62.
Kaseke, N., & Hosking, S. (2012, June). Cost of Load Shedding to Mines in Zimbabwe: Direct
Assement Approach. Retrieved April 10, 2016, from [Link]
doc/IJPSS_JUNE2012/[Link]
Kateregga, E. (2009). The welfare costs of electricity outages: A contingent valuation analysis
of households in the suburbs of Kampala, Jinja and Entebbe. Retrieved April 20, 2016,
from [Link]

Kazungu, I., Moshi, J., & Mchopa, A. (2014). POWER RATIONING DILEMMA: A BLOW TO SMALL
&
MEDIUM ENTERPRISES (SMEs). International Journal of Economics, Commerce and
Management United Kingdom, 1-14.
Kish, L. (1965). Survey Sampling. New York: Wiley.

46
Mason, M. K. (2009). Research on Small BusinessesMason. Retrieved December 16, 2016, from
[Link]

Ministry of Commerce, Trade and Industry. (2007). Small and Medium Enterprises Survey 2003
– 2004.

Retrieved April 15, 2016, from [Link]

Ministry of Finance. (2016). 2015 Annual Economic Report. Lusaka: Ministry of Finance.
Minnaar, U. (2015). Cost of unsaved energy. Johannesburg: SAIEE, Wattnow magazine.
Nuwagaba, A. (2015). Micro Financing Of Small and Medium Enterprises (SMEs) in Zambia.
Retrieved April 5, 2016, from [Link]
pdf
Nyamazana. (2015). Zambia Economic Cost of Load shedding. Lusaka: Institute of Economic
and Social Research, UNZA.
Nyasha , K. (2014). A Comparative Cost Assessment of Electricity Outages and Generation
Expansion in Zimbabwe. Retrieved December 2016, from [Link]
IJARMSS/Apr2014/[Link]
Pasha, H. A., Ghaus, A., & Malik, S. (1990). The economic cost of Power outages in the industrial
sector of Pakistan. Energy Economics, volume 4, pp. 301-18.

Project Gutenberg . (n.d.). Retrieved April 7, 2016, from [Link]

loadshedding
Samboko, P., Chapoto, A., Kuteya, A., Kabwe, S., Mukuka, R. M., Mweemba, B., et al. (2016,
March). The Impact of Power Rationing on Zambia’s Agricultural Sector. Retrieved April
6, 2016, from [Link]
Solomon, K. F., & Yao, L. (2015). Electricity Power Insecurity and SMEs Growth: A case Study of
the Cold Store Operators in the Asafo Market Area of the Kumasi Metro in Ghana. Open
Journal of Business and Managenment, 3, 312-325.

Terry. (2001). Cost of unsaved energy, World bank. Washington DC: Tata Energy Research

Company. The Overseas Development Institute. (2014). Electricity insecurity and SMEs .

Retrieved March 2017, from [Link]

media/57a089e1ed915d622c000441/61270- [Link]
Yao , L., & Solomon , K. F. (2015,). Electricity Power Insecurity and SMEs Growth: A Case Study
of the Cold Store Operators in the Asafo Market Area of the Kumasi Metro in Ghana.
Open Journal of Business and Management, 3,312-325.

Zambia Institute for Policy Analysis & Research. (2015). Electricity Sub-Sector Brief Quaterly
Report.

Lusaka: Zambia Institute for Policy Analysis & Research.

47
48
49
Head Office Copperbelt Office Livingstone Office Chinsali Office
Plot No. 9330, Mass Media Plot No. 332 Plot No. 708 Plot No. 76
Off Alick Nkhata Road, Independence Avenue Chimwemwe Road Mayadi
P. O. Box 37631, P.O. Box 22281 Nottie Broadie P.O. Box 480052
Lusaka, Zambia. Kitwe, Zambia P.O. Box 60292 Chinsali, Zambia
Tel: 260-211-258844 - 49 Tel: +260 212 220944 Livingstone, Zambia Tel: +260 214 565170
Fax: 260-211-258852 Fax: +260 212 220945 Tel: +260 213 321562-3 Fax: +260 214 565171
Fax: +260 213 321576

E-mail: erb@[Link] I Website: [Link] I Toll Free Line: 8484

50