Cost of Power Interruption

Generally, customers welcome timely replacement of ageing equipment, upgrading of facilities to

improve service, as well as expansion projects to meet growing demand. However, they are
understandably wary about the rate impact of said projects. Aside from determining the projected
costs, there is also a need to quantify the benefits of each significant project. A threshold amount needs
to be identified for this purpose.
For replacement projects, rate impacts should consider the offsetting of replacement depreciation vs
current depreciation. New facilities are likely more expensive than existing ones thus resulting in net
rate increase. However, improvement in technology, longer expected life, as well as efficiency gains in
manufacturing could also result in a negative net effect a reduction in depreciation expense.
Moreover, a positive growth rate of customer base also reduces the effective impact of additional
investments. If the amount of investments simply replaces current levels of depreciation while savings
are realized in transmission loss reduction and elimination of congestion events, the impact may even
result in a net rate reduction.
However, the preceding is easier said than done. Numerous data need to be gathered to arrive at
reasonable estimates of costs associated with new projects. Sources would include cost of awarded
projects broken down into specific details such as equipment cost per type of equipment per size, etc.;
construction costs; regulatory or compliance costs; right of way acquisition cost per region per type of
property; and so on. Statistics on transmission loss; power service interruption and cost impact;
congestion occurrence and cost impacts; etc. need to be compiled in order to come up with simulation
and calculation figures to justify the above projects. The same would be needed to justify budgets for
operation and maintenance, spare parts acquisition, etc.
The buy-in of these projects would not be possible without extensive public consultation. It would be
great if these meetings can be done on a more personal and per project basis so that more questions
can be entertained and issues clarified. Documentation of the discussions should be kept and can be
attached to the regulatory filings later on.
Transmission customers as well as electricity consumers should also be aware of what they are giving up
by preventing the expansion of the transmission of facilities to improve reliability as well as to meet load
growth. The cost of power interruptions include not only foregone revenues but also the cost of
repairing damaged equipment, cost of materials in process clean-up, manufacturing process start-up
costs, as well as loss of customer confidence. Several studies have been done to estimate the value of
lost load or the cost of power interruptions. While they do not agree on the exact cost and even on the
calculation methodologies, what is clear and common among them is that the value of lost load is much
higher than the actual price paid for electricity service. The following studies were used to analyze the
cost of power interruption in the Philippines:

1.

Understanding the Cost of Power Interruptions to U.S. Electricity Consumers

Kristina Hamachi LaCommare and Joseph H. Eto
Energy Analysis Department
Ernest Orlando Lawrence Berkeley National Laboratory
University of California Berkeley
Berkeley, California 94720
Environmental Energy
Technologies Division
September 2004
http://eetd.lbl.gov/ea/EMP/EMP-pubs.html

Our analysis shows that:

The majority of outage costs are borne by the commercial and industrial sectors;
As a result, although there are important variations in the composition of customers within
each region, the total cost of reliability events by region tend to correlate roughly with the
numbers of commercial and industrial customers in each region; and
Costs tend to be driven by the frequency rather than the duration of reliability events. Related
to this last finding, our work reveals the importance of short-term, momentary interruptions,
which last 5 minutes or less. Figure ES-2 shows that (more frequent) momentary power
interruptions have a stronger impact on the total cost of interruptions than (less frequent)
sustained interruptions, which last 5 minutes or more.

Table 10 shows the costs per outage per customer used in the Tobit regression equation to
calculate the total cost of power interruptions to U.S. electricity customers. The cost data are
classified by customer class (residential, commercial, industrial) and outage duration for the U.S.
The costs per outage per customer for individual regions are presented later in this report as
part of our sensitivity analysis. Our initial estimate uses the U.S. estimated cost per outage per
customer. This table represents the costs for momentary interruptions (i.e. 0 sec), 1-hour
outages, and the length of outage calculated from our trimmed means of SAIDI and SAIFI (i.e.,
Sustained Interruption.) Both the 0 sec and Sustained Interruption costs are used to
derive our initial estimate with the 1-hour cost used later in our sensitivity analysis. It is shown
here to indicate the costs associated with this commonly reported outage length. The sustained
interruption cost-per-outage-per-customer assumes the trimmed mean outage duration of 106
minutes presented in Section 4.2.

2.

Power Interruptions Cost Nation $80 Billion Annually

Berkeley Lab study focuses on state of U.S. power grid
By Robert Longley, About.com Guide
Dateline: February 2005
Another important conclusion of the study is that momentary interruptions, which are more
frequent, have a bigger impact on the total cost of interruptions than sustained interruptions,
which are less frequent. Momentary interruptions were responsible for two-thirds of the cost, at
$52 billion, while sustained interruptions of five minutes or more caused $26 billion. This
finding underscores that fact that, for many commercial and industrial customers, it is the length
of the down-time resulting from a loss of power that determines the cost of interruption, not
necessarily the length of interruption itself, according to Eto.

3.

Interruption Costs, Customer Satisfaction and Expectations for Service Reliability (May 1996)
Michael J. Sullivan, Vice President
Freeman, Sullivan & Co.
San Francisco, CA 94105 USA
B. Noland Suddeth, Manager of Transimission Control
Duke Power Company
Charlotte, NC 28201 USA
Terry Vardell, Manager of hdket ResearL-.
Duke Power Company
Charlotte, NC 28201 USA
3

Ali Vojdani, Manager of Power system Analysis

Electrical Power Research Institute
Pido Alto, CA 94304 USA

The above table also shows that customer outage costs are borne mainly by commercial and
industrial customers. Likewise, the cost per kWh at an average of $7.70 is much higher than the
cost of electricity in the USA, which is in the order of cents per kWh.
Customer expectations:
In terms of customer expectations, the two figures below indicate that residential customers are
more amenable to service interruptions for momentary as well as extended periods than
commercial customers are.

4.

The costs of electricity interruptions in Spain

Are we sending the right signals?
Pedro Linares, Luis Rey
This working paper has been developed within the Alcoa Advancing Sustainability Initiative to
Research and Leverage Actionable Solutions on Energy and Environmental Economics WP
FA5/2012.
The consequences of electricity interruptions can be classified into three categories (Tasgosz
and Manson, 2008):
Direct economic impacts:
Loss of production
Restart costs
Equipment damage
Raw material spoilage
Indirect economic impacts:
The cost of income being postponed
The financial cost of loss of market share
Social impacts:
Uncomfortable temperatures at work/home
Loss of leisure time
Risk to health and safety
In the literature, there have been different attempts to quantify electricity interruption costs.
The three most common methods are:
a.

Customer Surveys: In this method, surveys are employed to obtain information from
industrial, commercial and residential sector customers. The objective is to obtain a
direct or indirect valuation of interruption costs from customers. Direct approaches are
employed for those customers with a good knowledge of interruption consequences
(e.g., industrial sector and other large electrical users). Under some guidance,
customers are asked to identify the impacts and evaluate the costs related an electricity
interruption. When interruption impacts are less tangible and the monetary loss is more
difficult to evaluate, indirect evaluation methods are employed (e.g., for the residential
sector). Usually people are asked about their willingness to pay (WTP) to avoid
interruptions, or their willingness to accept (WTA) a compensation for having a higher
number of interruptions. As mentioned before, given that security of supply can be
considered as a public good, consumers will tend to overestimate their interruption
costs to free-ride on the system. On the other hand, they can also be interested in
underestimating them if their contribution to paying for the cost of security of supply is
higher than their share in the costs of an interruption.

b.

Case studies: Past events, as the blackouts occurred in California in 2001 and 2002, can
be used to quantify the cost of power interruptions. The advantage of this method is
that estimations are based on real events rather than hypothetical scenarios. It is easier
for electricity consumers to provide a more detailed cost evaluation when they have
experienced an interruption. However, this methodology is limited by the specific
5

characteristics of the outage studied (e.g., place, time, duration); and it is difficult to
generalize the results.
c.

The production function approach: This method uses the ratio of an economic measure
(e.g., gross domestic product, gross value added) and a measure of electricity
consumption (e.g., kWh) to estimate interruption costs by sector. The objective is to
find the value of one unit of electricity, also known as the Value of lost load (VoLL). For
example, if the gross value added of a sector is 10 million using 1 million kWh of
electricity, the cost of a power interruption would be 10/kWh. Under the production
function approach, it is assumed that electricity is essential for production, which is not
always true. In some sectors, an electricity interruption does not necessarily imply a
production break. Furthermore, production may be postponed or displaced to other
locations or time slots. Therefore, this method may overestimate electricity
interruption costs. It is also easy to notice that this measure of VoLL is the inverse of
energy intensity, which measures the amount of energy required to produce a unit of
economic output. Therefore, electricity-intensive sectors will by definition show a lower
VoLL. However, this does not include the dependability of the sector on electricity
consumption, which may also be very relevant for the quantification of the cost of lost
load.
The table below summarizes estimated cost of lost load in different countries using
various methodologies.

The production function approach, when applied to Spains electricity use, resulted in
the table below. The detailed available information allowed the disaggregation of the
VoLL into various sectors like Agriculture, Manufacturing, etc.

5.

Philippines: Value of Lost Load (VoLL) Using Production Function Approach

In order to estimate the value of lost load (VoLL), or the cost of unserved energy (or cost of
power interruption) using the production function approach, data was gathered from various
government agencies. Power statistics from the Department of Energy (DOE) were available
from 2003 to 2011 for the industrial, commercial and other sectors (Power-Statistics-2011.pdf).
From the National Statistical Coordination Board (NSCB), statistics on Gross Domestic Product
and Gross Value Added (GVA) by industry origin from 1998 to 2010 were also obtained
(Quarterly_Summary_93SNA_series_1998_2010_NSIC.xls).

YEAR
2003
2004
2005
2006
2007
2008
2009
2010

GDP
@Current
Prices
PMn
4,548,102
5,120,435
5,677,750
6,271,157
6,892,721
7,720,903
8,026,143
9,003,480

Philippines
Electricity
Sales
MWh
42,719,994
44,075,959
45,159,402
45,672,173
48,009,038
49,206,114
50,867,841
55,265,769

Philippines
VoLL
P/kWh
106.46
116.17
125.73
137.31
143.57
156.91
157.78
162.91

Table 1a Value of Lost Load or Cost of Power Interruption

7

GDP (less AHFF)

@Current Prices
YEAR
2003
2004
2005
2006
2007
2008
2009
2010

PMn
3,970,297
4,439,139
4,958,674
5,495,469
6,031,356
6,698,389
6,976,269
7,894,762

Philippines
Electricity
Sales
MWh
42,719,994
44,075,959
45,159,402
45,672,173
48,009,038
49,206,114
50,867,841
55,265,769

Philippines
VoLL
P/kWh
92.94
100.72
109.80
120.32
125.63
136.13
137.14
142.85

Table 1b Value of Lost Load (excluding AHFF)

In 2010, the cost of power interruption based on GDP was P162.91/kWh (Table 1a). Adjusted
downwards to exclude Agriculture, Hunting, Forestry, & Fishing (AHFF), the figure is still quite
significant at P142.85/kWh (Table 1b).
In Table 2a, GVA for the industrial sector included Mining & Quarrying, Manufacturing,
Construction as well as Electricity, Gas and Water Supply. The commercial sector GVA was taken
from the services sector which included Transport, Storage & Communication; Trade and
Repair of Motor Vehicles, Motorcycles, Personal and Household Goods; Financial
Intermediation; Real Estate, Renting & Business Activities; Public Administration & Defense,
Compulsory Social Security; and Other Services. The industrial and commercial electricity sales
were taken directly from the DOE statistics. Unfortunately, electricity use has not been broken
down into subsectors as classified by the NSCB. This prevents the calculation of VoLL based on
similar classifications.

YEAR
2003
2004
2005
2006
2007
2008
2009
2010

Gross Value Electricity Sales

Industrial
Added
to Industrial
VoLL
Industrial
Customers
PMn
MWh
P/kWh
1,571,866
15,188,293 103.49
1,728,281
15,011,992 115.13
1,921,019
15,705,160 122.32
2,100,382
15,888,141 132.20
2,278,254
16,521,717 137.89
2,538,461
17,030,903 149.05
2,545,104
17,084,427 148.97
2,932,279
18,576,307 157.85

Gross Value Electricity Sales

Commercial
Added
to Commercial &
VoLL
Commercial Other Customers
PMn
MWh
P/kWh
2,398,431
12,175,118
196.99
2,710,858
13,143,730
206.25
3,037,655
13,422,884
226.30
3,395,087
13,953,831
243.31
3,753,102
15,111,731
248.36
4,159,928
15,530,981
267.85
4,431,165
16,279,670
272.19
4,962,483
17,856,454
277.91

Table 2a Cost of Power Interruption for Industrial & Commercial Customers

The average Meralco residential electricity rate in 2010 was around P10.50/kWh
(typical_consumption_levels_January2010.pdf
&
typical_consumption_levels_December2010.pdf).
Industrial and commercial customers typically pay at least 10% less due to lower transmission,
8

distribution, and system loss charges (summary_schedule_rates_December2010.pdf). On the

other hand, the values of lost load for industrial and commercial customers are P158/kWh and
P278/kWh, respectively while the total for the country comes up to around P143/kWh
(adjusted). Unfortunately, the dearth of statistics for residential customers prevents a similar
calculation for the cost of power interruption for residential customers. It is interesting to note
that, like Spain, the cost of interruption for industrial customers is lower than that for
commercial customers, by about 44%, even if industrial consumption is slightly higher than
commercial consumption of electricity.
REGION
LUZON

ITEM / YEAR
Family Income (PMn)
Residential Sales (GWh)
VoLL (P/kWh)

2003
1,814,320
11,796
154

2006
2,207,141
11,802
187

VISAYAS

Family Income (PMn)

Residential Sales (GWh)
VoLL (P/kWh)

394,544
1,785
221

502,943
2,036
247

662,070
2,341
283

MINDANAO

Family Income (PMn)

Residential Sales (GWh)
VoLL (P/kWh)

424,057
1775.703
239

535,639
1,992
269

688,233
2,362
291

Philippines

Family Income (PMn)

Residential Sales (GWh)
VoLL (P/kWh)

3,245,723
15,830
205

4,079,310
17,504
233

2,632,921
15,357
171

2009
2,729,007
12,801
213

Table 2b Cost of Power Interruption for Residential Customers

To estimate the cost of power interruption for residential customers, the statistics on household
income was used (TABLE 1 Number of Families and Family Receipts by Income Class and By
Region and Source of Receipts, 2009.pdf).

REGION
LUZON

ITEM / YEAR
Regional GDP ('000 Pesos)
Electricity Sales ('000 kWh)
VoLL (P/kWh)

2009
5,828,519,226
37,859,434
153.95

2010
6,557,990,226
41,388,794
158.45

2011
7,061,837,093
41,706,246
169.32

VISAYAS

Regional GDP ('000 Pesos)

Electricity Sales ('000 kWh)
VoLL (P/kWh)

1,012,335,298
6,309,113
160.46

1,127,164,171
7,036,059
160.20

1,239,890,561
7,224,369
171.63

MINDANAO

Regional GDP ('000 Pesos)

Electricity Sales ('000 kWh)
VoLL (P/kWh)

1,185,288,864
6,699,295
176.93

1,318,326,094
6,840,916
192.71

1,433,793,389
7,166,977
200.06

Philippines

Regioinal GDP ('000 Pesos)

Electricity Sales ('000 kWh)
VoLL (P/kWh)

8,026,143,388
50,867,842
157.78

9,003,480,491
55,265,769
162.91

9,735,521,043
56,097,592
173.55

Table 3 Regional Value of Lost Load

On a regional basis, it would seem that the cost of interruption in Visayas and Mindanao are
higher than that in Luzon. It is significant to note that the cost of power interruption in
Mindanao in 2011 is around 15% higher than the national average. Unfortunately, Mindanao
has been experiencing frequent and prolonged power interruptions for the past several years
already!
This method of estimating VoLL or cost of power interruption excludes the cost of damaged
equipment as well as cost of production startups attributable to power interruptions.
Moreover, when interruption occurs during daytime and on weekdays, the cost of power
interruption is significantly higher since as significant share of overall GDP or GVA is produced
during daytime.
On the other hand, the calculations do not cover the possibility of shifting the use of electricity
to a future time to recover from the interruption, albeit at an additional cost too. Moreover, as
mentioned earlier, the amount (in Table 1a) is overstated by the fact that agricultural and
fisheries production may be less sensitive to the supply of electricity. For example, in the
regional determination of VoLL, it is not clear whether agricultural production will actually be
interrupted when electricity is interrupted. How much this industry is dependent on electricity
is not yet very clear. Nevertheless, the prevailing average cost of electricity (around P11/kWh) is
significantly lower than even 10% of the VoLL. This is likely the reason why many commercial
establishments have installed small standby generator sets in order to ensure access to
continuous supply of electricity especially during the years of frequent power interruption!

References:
1. Understanding the Cost of Power Interruptions to U.S. Electricity Consumers, Kristina Hamachi
LaCommare and Joseph H. Eto, Energy Analysis Department, Ernest Orlando Lawrence Berkeley
National Laboratory (55718 understanding cost of interruption.pdf).
2. Power Interruptions Cost Nation $80 Billion Annually, Berkeley Lab study focuses on state of U.S.
power grid, By Robert Longley (Power Interruptions Cost Nation.docx).
3. Interruption Costs, Customer Satisfaction and Expectations for Service Reliability (May 1996),
Michael J. Sullivan (00496185 ieee interruption cost.pdf).
4. The costs of electricity interruptions in Spain, Are we sending the right signals? Pedro Linares,
Luis Rey (WPFA05-2012.pdf)
5. Philippine Power Statistics 2011 (Power-Statistics-2011.pdf).
6. Philippines GDP & GVA (Quarterly_Summary_93SNA_series_1998_2010_NSIC.xls).
7. 2010 Meralco Jan & Dec Residential Power Rates (typical_consumption_levels_January2010.pdf
& typical_consumption_levels_December2010.pdf).
8. 2010 Meralco Indl & Coml Power Rates (summary_schedule_rates_December2010.pdf).
9. 2003 Household Income (TABLE 1 Number of Families and Family Receipts by Sources of Receipt,
by Income Class and by Region 2003.PDF).
10. 2006 Household Income (TABLE 1 Number of Families and Family Receipts by Sources of Receipt,
by Income Class and by Region 2006.pdf).
11. Household Income (TABLE 1 Number of Families and Family Receipts by Income Class and By
Region and Source of Receipts, 2009.pdf).
12. Calculation worksheets (value of lost load.xls)

10