Measuring the socio- economical returns of e- Government: lessons from egep First LOG-IN Africa Methodology Workshop, 8 10 June 2006, Tangier Morocco Dr. Andrea Gumina, PhD Project Leader, egov@luiss - Luiss Business School CEO, NExTT a.gumina@methisgroup.it 1
Definition of e-government egovernment refers to a deep ICT-enabled transformation process, which could take place within the Public Sector and with reference to its relations with stakeholders. Such a process is the outcome of a correct mix among re-organisation, human capital empowerment and ICT solutions. The main goal of the egovernment is to act on effectiveness and efficiency of Public Administrations, while an indirect objective is to transform the entire corpus of relationship within a Knowledge Society.To do so, a number of mechanism should be put in place, in order to assure the diffusion of the induced innovation and technical progress at all levels. It stems from this, that a number of activities should be put in place by Public Sector: some are regulatory, some others refers to policy issues. They should all purse to make progress generated within the Public Sector, real and lasting at a Society-level. As a result, egovernment should act as a multiplier on productivity and growth, becoming a powerful driver for socio-economic development. 2
Measuring the main aims of LOG-IN Africa Three issues of LOG-IN Africa seem to imply a measurement effort: It will assess the current state and outcomes of e-local governance initiatives in Africa, and in particular how ICTs are being used to realize good local governance at four levels: a) the internal organizational processes of local governments; b) the provision of information and service delivery; c) the promotion of the principles of good governance; and d) public participation and consultation. It will generate modular outcome assessment framework, with specific indicators linked to a conceptual framing of the characteristics of good governance, will be developed. Finally, the Network will generate research findings contributing to more effective policy making and implementation in e-local governance in Africa. Main fields to measure refers to socio-economic impact of egovernment, as well as to the investment returns of the investment made / to be made 3
A Socio-Economic View of e-government makes measurement mandatory Scarce financial resources, needs to be effective in policy making and to produce value for stakeholders, make measurement and monitoring activities mandatory. Furthermore, they are going to be considered as investments themselves, rather than as expenses. 4
5 The egep Model: basic tenets Smith s Effect Smith s Effect Labour productivity Investment Led Effect Investment Led Effect Take-up Effect Take-up Effect Back-office Effect Back-office Effect Ricardo s Effect Ricardo s Effect ( ) π PS t jn t tp jn t cp jn sw t jn hw t t t t PS n t egov n t PS PS t PS li rs ht f E E E E e ASVU ASCU L d P W c by ) (,,,, 1,,,, = π
Economic Model Efficiency Tangible financial gains Measurable side of back-office effect egov investments Effectiveness Combined efficiencyeffectiveness: Increased output New Services Increased User Value Increase in pub. sect. productivity Increase in pub. sect. Share of GDP GDP GROWTH Take up & connectivity egovernment leads to increased public sector labour productivity, which in turn, contributes to a number of intermediate outcomes (better services, cost savings, etc), and to GDP growth 6
The Smith s Effect In the private context, the Smith effect takes into account how the growth of market/specific sector influences the labour productivity. The adaptation to the Public Sector is based on the assumption that the introduction of egovernment increases the supply capacity of the Public Sector and that all services produced are also demanded by users. 7
The Ricardo s Effect In the private context, the Ricardo s effect shows how an increase in the spread between wages and the price of technology (machines) pushes businesses to gain productivity, through a substitution process between technology and employees The adaptation to the public sector implies a reformulation: when the cost of innovation compared with that of personnel decreases, it can be efficient to partially substitute and partially integrate the latter with a wide implementation of egovernment services 8
The Schumpeter s Effect The characteristic element of this component is the consideration of ICT investments and related aspects: consulting, training, hardware, software, etc. The impact of such innovations (as new innovations replace the older ones) generate an increases in productivity, after a time lag 9
Other Effects Back-office reorganisation effect is connected to the difference between the cost of delivering a given service and its perceived value. If this cost grows, it is necessary to modify the delivery organisation influencing by this way productivity Take-up effect : the more the social environment is receptive (e.g. because of a wide ICT diffusion, or because of a broad, deep- rooted use of e- services) the more egovernment increases public sector productivity 10
Model testing by macro variables Economic Model ALP = Average Labour Productivity PS = Public Sector Share Variables output /full-time equivalent employees general government output as % GDP Sources Eurostat Eurostat Smith s Effect Y= Market Dimension Ricardo s Effect W = Cost of staff K = Cost of capital equipment Schumpeter s Effect H = Cost of hardware S = Cost of software C = Cost of consulting T = Cost of Training Investment Effect I= Investment general government output general govt. employees compensation price index of gen. govt. goods investments Estimated expenditure Estimated expenditure Estimated expenditure Estimated expenditure gross fixed capital formation of gen. govt. Eurostat Eurostat Eurostat egep Eurostat 11
Two versions tested Two forms of our model: one based on growth rates (labelled A ), and one based on absolute values ( B ) A. ALP = a by cw dk ei fegov B. ALP = a by cw dk ei fegov Based on data for 20 countries: Austria, Belgium, Czech Republic, Denmark, Finland, France, Germany, Greece, Hungary, Ireland, Italy, Netherlands, Norway,Poland, Portugal, Slovakia, Slovenia, Spain, Sweden, United Kingdom; data for the period 1999-2005 12
Results (sign of coefficients) Effects Variabl e Equation A Equation B Smith s Effect Y * * Ricardo s Effect W K * - * Schumpeter s Effect I egov * * * * (*) Coefficient significantly different from 0 at 95% confidence level Note: a) Constant prices variables (Y,I, egov) expressed in millions of Euro, baseline year 2000. b) Equation A was fitted by a Fixed-Effects Panel model (68 observations), Equation B was estimated with the Arellano-Bond technique (68 observations). 13
Model Projections Impact of egovernement egovernment on Average Labour Productivity (ALP) 2005 2006 2007 2008 2009 2010 A 1.93% 0,69% 1.67% 0,73% 2.04% 0,80% 0,00% 0,35% 0,00% 0,16% 0,00% 0,07% B 1.44% 0,76% 1.39% 0,62% 1.70% 0,59% 0,31% 0,09% 0,14% 0,05% 0,06% 0,03% 1,00% 0,90% 0,80% 0,70% 0,60% 0,50% 0,40% 0,30% 0,20% 0,10% 0,00% Graph 2.1 Projected GDP growth 2005 2006 2007 2008 2009 2010 dgdp B dgdp A Cumulative contribution to EU-25 GDP growth within 2010 A : 2,06% B : 1,36% (with e-procurement, without back-office and take-up effect) 14
Main problems about data relevation Measuring perfomance recquires fresh and adequate micro-data Data collecting should be easy and cheap for PAs, in order to be as frequent as possible Data should be collected and processed with a strong attention to the local level Processing methodology should be clear and agreed among Public Sector s representatives. 15
Main findings and policy reccomendations from the egep methodology Increase both the efficiency and the effectiveness of egovernment inside Public Administrations with reference to their main stakeholders (policies 1 to 4); Promote the dissemination of specific tools to measure performance, together with a favourable cultural background (policy no. 5); Set-up a number of complementary policies, in order to foster the take-up effect, and finally promote productivity and growth. 16
Policy no. 1: To share egovernment goals Focusing every egovernment project on front-office as well as on back-office, paying much attention both to the deployment of easy-to-use services and to the setting up of productivitysensitive policies, such as those for general and specific training or re-organisational issues. Promoting a favourable scenario, where social, economic, financial, legal and institutional issues play a major part in overall local growth when adequately oriented. Defining an operational measurement framework, which also favours the establishment of comparative benchmarking and prizing processes. 17
Policy no. 2: To guide egovernment towards shared objectives. The promotion of a horizontal egovernment. A review of egovernment vertical policies. Specific principles, linked to technical features. The promotion of both a user-centred and a locally-aimed egovernment. Due to budget constraints, egovernment policies should not be oriented only to local growth, but should also take into account user needs. As a consequence, further implementation should be based upon a number of subjective and objective priorities, combining the need to satisfy user needs and the opportunity of realising programmes which refer to the true needs and potential of local areas, such as macroeconomic entities. This would lead to a progressive deployment of egovernment projects, following different priorities in different areas, maybe not homogeneous, thus giving measurement, monitoring and fine tuning activities a sharper and more effective meaning, in the light of promoting sustainable and lasting local growth. The promotion of specialisation within the Public Sector. Horizontal and pervasive egovernment should benefit from a higher degree of local specialisation within different branches of Public Administration (if accompanied by perfect interoperability), rather than from an indiscriminate upgrade of organisational issues within the entire Public Sector. However, in general, the size of single may find it more useful to identify and promote such economies internally, smaller administrations may find outsourcing to be a smarter and more suitable solution. 18
Policy no. 3: To favour a performancefriendly environment within the Public Sector Accelerating the shift from evaluation to performance rewarding. Together with a mandatory measurement approach, performance should be linked to a prizes and penalties mechanism. Public Administrations do not have strong incentives to set particularly ambitious objectives and to try to achieve them. This could lead to vicious circles: an adverse selection mechanism could take place, where establishing goals that are too ambitious could mean risking not reaching them. The only way to break these vicious circles is to ensure strong financial and career incentives linked to performance for both personnel and public managers. Working against motivational crowding out. Fostering the adoption of reforms that encourage performance measurement. Finally, such a framework requires regulatory and legislative ad hoc reform promotion: this refers, for instance, not only to the above-mentioned prize and penalty mechanism, but also to guaranteeing mandatory performance measurement, and linking legitimisation of policy makers decisions to e-democracy tools. 19
Policy no.4: To create a friendly financial framework for egovernment Let Administrations invest, if useful. Reward the best Administrations and Organisational Units. Invest Cost Savings. 20
Policy no. 5: To make performance measurement mandatory Define a homogeneous road-map. The entire Public Sector will certainly not increase its egovernment readiness unanimously; however, it is necessary to define a clear and homogeneous road-map for such a complex transformation, with precise features and goals. Otherwise, the general enhancement of the Public Sector will be poor and egovernment could be jeopardized. Moreover, if performance measurement aims to affect the budget (negatively or positively), more attention should be paid to the definition of a general benchmark. Define a clear, homogeneous and simple measurement framework. Measurement tools should be easy to handle and use, but also as complete and effective as possible, in order to minimize any coordination cost that may arise. Furthermore, as more e-services gradually become ICT-enabled, the administrative burden to measure use, efficiency and satisfaction should decrease. In this case, even a limited measurement effort, in absolute value, could be considered a substitute for the introduction of a more advanced egovernment. Local matters. We believe that egovernment will spread its effects on growth and performance especially at local level, and that the measurement framework should take this feature into account: an adequate observation scale should be achieved (i.e., NUTS3 areas). Indeed, internal measures could be aggregated in order to allow a comparison with external effects: this is a must when a prize/penalty framework is to be set up. Measure the effort and the gap. History matters, especially if budgetary cuts or gains depend on performance measurement. In this light, it becomes compulsory not only to take into account the effort and the results, measuring the period-by-period differences, but also to bear in mind the remaining gap and the absolute values of such efforts. Otherwise, the observation of political and administrative commitment may be distorted. 21