Economic Valuation of Electricity Supply Security – Why is it important?
Ensuring the continuity of supply is essential for any advanced economy to function economically, socially and politically. Europe has enjoyed a high degree of electricity supply security (ESS) during the last few decades. However, the need for action to ensure high levels of ESS in the future is increasing, mainly because electricity production and distribution are currently undergoing significant restructuring.
While developing the necessary measures to secure the electricity grid and future supply is primarily a challenge for the engineering disciplines, it is the task of economic research to support the development of a system of incentives to counterbalance possible market failure and therefore enable the implementation of societally optimal technical measures. One central prerequisite for developing an efficient regulatory system is quantifying the value of electricity supply security.
As ESS constitutes a non-market good, and can be purchased only in combination with the physical product (electricity), the value of supply security cannot be determined directly. That is why usually the failure of electricity supply, and in particular the cost of power cuts, is used to assess the value of supply security (see bibliographic references ,  and ).
Elaborated economic assessments of the value of electricity supply security are essential especially in the discussions regarding necessary investments for maintaining and upgrading the current transmission and distribution infrastructure. The tool provides a multifunctional “easy to use” web tool supporting for example infrastructure decisions and energy policy. Serving a variety of interested users, quantitative assessments such as the value of lost load, the outage related damage of each economic sector and households in every country of the EU at the NUTS 2 level, as well as of the energy not supplied in the case of any user defined blackout, can be conducted.
In order to model the economic costs of widespread power outages, losses due to electricity outages are typically classified into three categories (bibliographic reference ):
- Direct costs
- Indirect costs
- Resulting long-term costs of macroeconomic relevance
In the public eye direct economic losses are usually at the top of the list. They are a direct result of the failure, e.g. repair costs for defective electrical infrastructure facilities. However, direct economic losses are usually limited and subordinate to indirect economic losses. These indirect costs also arise in direct connection with the failure, yet they belong to that part of the total losses resulting from the absence of electricity supply in the aftermath of the failure. Examples are the loss of productive activity, or lost value added. Through multiplier effects due to the marked dependence of some industries on the flawless functioning of other economic sectors, these indirect costs make up a significant proportion of the total costs (bibliographic reference ).
The analysis of the damages inflicted on businesses (non-households) in the case of power interruptions relies on two different methods. On the one hand, production data on the gross value added of businesses, industry and public administration were incorporated as a central indicator of economic activity. This is because economic activity is in most cases very closely connected to electricity supply.
The second part of the assessment of non-households is based on a comparison of typical damages per kWh not supplied in certain industries and sectors. This value of lost load (VoLL) approach assigns every unit of electricity not supplied a damage for a certain country and sector. The different VoLL values were regressed on the characteristics of the outage under scrutiny. The VoLL influencing 4 factors were incorporated into the model to assess a broad variety of power outages which are not restricted to the characteristics of the original power outage settings .
Finally, the economic losses of individual non-households were clustered to make it possible to form complete aggregates of sectors or regions subsequently. A key factor is the utilization of control variables for holidays, weekends and after work hours. This implies for example, that a grocery's daily value added is considerably higher on workdays than on public holidays. However, damages occur on holidays as well (e.g. the lack of cooling and security appliances), which is accounted for in the model as well .
For a comprehensive analysis of the household sector it is necessary to represent non-monetary effects as well as material losses. Thus, an unprecedented household-level survey with a total of 8,336 participants (250-300 per country) was conducted to evaluate European households’ WTP to avoid power cuts. The econometric assessments which yielded the inputs for the tool are state of the art in the economic science.
RESULTS and APPLICATION
Being a main contribution of SESAME, this recent software development provides a tool for placing a value on the non-market good of “electricity supply security” for all member states of the European Union. Although the supply of electricity is relatively reliable in Europe, maintaining this degree of reliability in the future involves a number of challenges. Efficient decisions on investing in infrastructure are possible only if the value of supply security is determined. To obtain an objective result, a household survey unprecedented in scope and dimension (over 8,000 households) as well as an assessment of the macroeconomic costs to businesses, institutions and governments in the case of power outages were conducted.
Using this comprehensive approach to calculate the monetary value of a reliable supply of electricity for every NUTS 2 region (in most cases provinces or regional states), a fairly fine-mesh quantification of all economic sectors is possible. As a result, not just particularly vulnerable sectors (such as the semiconductor industry, papermaking or data-generating processes), but all sections of the economy as per NACE 2008 are modeled. The wide range of possible blackout scenarios, lasting from one to 48 hours, covers many different conceivable outages for most of the provinces of the EU (266 in total); it is thus possible for the first time to judge subsectors of the European economy province by province as regards their degree of dependence on a reliable supply of electricity. Assessing households’ vulnerabilities in the case of power outages, the economic costs of businesses and the quantity of electricity not supplied in the case of widespread power outages is crucial for future policy decisions, investments in utilities, negotiations between regulatory bodies and grid operators, etc.
The tool incorporates data on businesses’ power outage interruption costs and joins them with the largest data set ever compiled on willingness to pay for uninterrupted power. The maximum blackout time supported by the tool is capped at 48 hours. Longer power outages would exhibit hard-to-assess social impacts and are thus omitted. Outages in the second to minute range, which are potentially relevant, but mostly heterogeneous in nature, are also nearly impossible to represent objectively in economic terms.
The costs due to damage to or the destruction of electricity infrastructure have for comparability reasons not been included into the model, as the model simulates the resulting blackout, independently of what caused it .
The model depicts the effects of the simulated blackout in all assessed 266 provinces of the EU and for nine economic sectors, as well as for households in the region selected for the scenario in question. To do this, several indicators of the economic impact of the widespread blackout investigated on non-households and households are calculated. Economic loss and especially the electricity shortfall due to the outage are used as the key indicators in the case of non-households and are derived from the synthetic load profiles of the various sectors. Furthermore an indicator used is the damage per unit of electricity not supplied (VoLL).
As an example of a very prominent power outage, namely the large-scale supply interruption which occurred in Italy September 28th 2003 is investigated by means of the assessment model. Figure 1 depicts the characteristics of this power outage 
Figure 1: Power Outage in Italy on September 28th, 2003 lasting from 3 to 16 hours
Based on the economic activity of the uninterrupted state of the European electricity system, the tool calculates the share of that activity being lost during a blackout. Additionally to the geographical differentiation, economic activity is split between 9 commercial and one household sector. Figure 2 presents the chain of restoration, which can be specified by the duration of the power outage in any of the relevant areas (Nuts II is in most European countries based on state borders).
Figure 2: Chain of Power Restoration after the Supply Interruption as selected in the tool
Table 1 presents the summary of this power outage which affected over 55 m. persons and lasted 3 hours in the north, 9 hours in the center, 12 hours in the south and 16 hours in Sicily.
In the case of the outage analyzed in this scenario, the whole of Italy except Sardinia is affected by a blackout lasting 3 to 16 hours. As a result the economic assessment of the losses and effects due to an outage of this kind is presented in Table 2.
As the scientific evidence on electricity supply security and its socio-economic dimension is facing growing public interest, there is a need for more research that puts a monetary value on “supply security”, particularly at the transnational level. Given that European markets for electricity are increasingly interlinked, and that interdependence across borders is more and more marked, there seems to be a very strong case for assessing “supply security” uniformly throughout Europe.
The interaction of supply security and the implementation and adaption of regulatory frameworks is among one of the most relevant issues with regards to a sustainable, secure and affordable supply of electricity.
For more information, please visit www.blackout-simulator.com
 However, processing the data of this study has revealed that the VoLL in different studies varies considerably within any given industry. Thus it was important to exclude extreme outliers and install an artificial boundary to prevent the coefficients from the VoLL database to lead to negative values for certain industries.
 In total nine economic sectors were incorporated into the analysis, all of which are based on the “NACE Rev. 2” system.
 These damages (while present in most power outage cases) are excluded from the assessment, since these losses depend on the cause of the outage (e.g. break in supply line, operator error or software problems) and are extremely heterogeneous.
 The outage cost calculation was restricted to the territory of Italy. Parts of Switzerland and states such as San Marino, which were also affected, were not included.
To develop the economic activity, data of European or national sources was exploited. Thus, the statistical units for which the economic activity is reported are limited by the availability of data.
 B. E. Baarsma and J. P. Hop, "Pricing power outages in the Netherlands," Energy, vol. 34, no. 9, pp. 1378-1386, 2009.
 M. de Nooij, C. Koopmans and C. Bijvoet, "The value of supply security. The costs of power interruptions: Economic input for damage reduction and investment," Energy Economics, vol. 29, pp. 277-295, July 2007.
 C.-K. Woo and R. L. Pupp, "Costs of service disruption to electricity consumers," Hong Kong, 1992.
 M. Munasinghe and A. Sanghvi, " Reliability of Electricity Supply, Outage Costs and Value of Service: An Overview.," IAEA Special Issue Electricity Reliability Issue, no. 9, 1988.
 P. a. F.-D. M. a. G. L. A. a. K. T. Centolella, "Estimates of the Value of Uninterrupted Service for The Mid-West Independent System Operator," 2006.
 Commissione di Indagine, "Black-out del sistema elettrico italiano del 28 settembre 2003 Rapporto della Commissione di Indagine," Rome, 2003.