The StatRes project - more performance transparency in the public sector Morten Strømgren Workshop on Measurement of non-market output in education and health October 3, 2006

Situation: Significant performance gaps in Norway Need to improve public-sector performance in Norway… - performance transparency is important "driver" EXAMPLE EDUCATION In school (grade 8-10), spending per pupil is 40 % above OECD average International comparisons show that skills are only average EXAMPLE HEALTH (1) No. 2 in OECD in health expenditure per inhabitant Life expectancy is lower in Norway than in countries spending much less - such as France, Italy and Sweden EXAMPLE HEALTH (2) Along with increased spending in our hospitals, cost efficiency is reduced The effect is even stronger if the spending increase is high On average, for each 1 % extra spending, production increases only 0,3 %

Challenge: Many information systems, low availability Limited access to data Presented in many ways Lack of comparability Low impact on performance "Bottom-up" - Detailed systems, input focus "Top-down" - Basic info. web sites, activity focus Impact Publicity Different information systems covering government activities National accounts Sector-wise web sites National statistics Examples:

Example (1): The KOSTRA system for municipalities Municipality: ABCDEFGHIJKLMNOAVERAGE Example: Cost benchmarking Performance indicators and measures for spending, activities, output, outcome Covers all 431 municipalities Updated annually in March, plus one revision Cost per home-help receiver in Oslo townships (NOK, 2005)

Example (2): Performance of hospitals in SAMDATA Performance indicators for health enterprises Spending, through-put, availability and efficiency Benchmarking among health enterprises and hospitals Example: Operating costs per stay and per day Health enterprise (hospital) Sykehuset Østfold Sykehuset Asker og Bærum Akershus universitetssykehus Aker universitetssykehus Ullevål universitetssykehus Operating costs per stay, adjusted Level 2004 Change Relative cost level Operating costs per day Level 2004 Change Relative cost level

Example (3): Results from schools in Skoleporten Performance measures for higher and lower secondary schools Levels: School, municipality, region, national Actively used by schools in efforts to improve Share of applicants for higher secondary school admitted by their first choice among the schools STATISTICS Share of pupils completed higher secondary school on schedule Share of pupils completed h. sec. school on "overtime" Share of pupils dropped out of higher sec. school Through-put School factsResourcesLearning envir.Learning results Example: Through-put measures

Analysis: Fragmented information, need for easier access OutcomeInputActivityOutput Low access (internal data) High access (Internet, easy) Medium access (Internet, hard) Internal data in sector/agency National accounts Various research projects National budget Office for national statistics Agencies' and sectors' information portals StatRes Mapping of some important sources of data:

StatRes: Web site showing performance indicators First preliminary draft

StatRes: Features… what's special? 1.Operated by Statistics Norway, ensuring independence 2.Indicators are defined in co- operation with sector specialists 3.Common framework for central government, linking 'input' with 'output' and 'outcome' 4.Accessible for everybody, not only experts 5.Allowing for comparisons, also internationally Input Activity Output Outcome ABC

Example: Possible indicators, pilot "higher education" Input Activity/unit Output Outcome Costs Employees Man-years Students Average annual student progress PhDs in progress Degrees PhDs completed Papers (rated) Grades? Ranking Alumni employment Average grades of applicants from higher sec. school Alumni salary compared to total cost of degree Quality ranking over spending ranking (e.g. share of GDP) KEY FIGURES KEY RATIOS Cost per student per year Students per academic man-year Share of exchange students Share academic man-years Cost per academic man-year Share of non- govt. financing Cost per degree Share graduated on time Papers per academic staff

Final remarks: Effects, assumptions More attention to public spending and results (output, outcome) Speed up the use of performance indicators in public sector Improve desicions, resource allocation, motivation - better performance Anticipated effects Assumption: strong efforts needed Most countries in the same situation: much relevant data, low accessiblity Strong efforts needed to make performance indicators and data available to a larger audience - increase use and impact

Thank you

Back-up slides Will not be used, unless detailed questions

Outline: Performance, transparency, the StatRes project 1.Performance gaps, examples 2.Need more transparency, examples 3.One approach: The StatRes project 4.Some challenges and our response Public-sector performance

Possible approach: Efforts along two dimensions Performance transparency Measure performance Display results Comparability Incentives and follow-up Build right competence Structures, technology etc. Support to agencies in performance improvement

Now establishing new system 'StatRes' for central government Ministry of Government Administration and Reform (idea, financing) Statistics Norway (system development) What? Why? Who? Need easily available fact base to evaluate effects of public spending, on various levels Internet publishing of statistics and indicators for central government sectors - based on existing data, not new reporting Developing framework, recruiting pilot areas, planning implementation Pilot areas and system design starting next month When? Planned lauch in 2007, thereafter include more sectors continuously How?

Developed from 1995 and onwards Covering all 434 municipalities Annual publication on 4 levels - Counties - Municipalities - Townships (Oslo) - Institutions (secondary el. schools) Covers inputs, activities, outputs, some outcomes Challenges: - Counts expenditures, not costs - Not very user friendly - No direct link inputs - outputs (2) Share of elderly at nursing homes that have single rooms (1) Unit costs (index) - municipal services KOSTRA system for municipalities on activities and output Brief facts - KOSTRA Kinder- garden Total Elderly care Primary school Municipality BCD Examples of applications EA

Effects of StatRes, some questions We believe StatRes will… -attract more attention to the use of public funds, and the results (output, outcome) -speed up the use of performance indicators in public sector -provide a common framework for assessing the results of public spending -be a tool used by both politicians, managers, employees and citizens -help improve desicions, resource allocation and performance in public sector Anticipated effects Questions we have asked ourselves Wait for better measures (costs, indicators, etc.) before displaying results on web? How to select relevant performance indicators, avoiding unintended behavior? Much overlap with existing sector systems for displaying performance? How to ensure a sustainable system somehow beyond political control? Who will be the users of StatRes - politicians, managers, employees, citizens…? Are parts of the public sector not suited for the input-output-outcome logic? Should StatRes include all public spending, also transfer payments to private sector?

Final remarks: Key questions, assumptions How to ensure a sustainable, consistent, credible and relevant system? How to avoid a sense of competition from existing performance-information systems? Which government expenditures should be included? Key questions Statistics Norway develops and operates the system Close co-operation with sectors in defining indicators Loyalty to a fixed framework, trying to link spending and results (transformation model, COFOG) Stressing that StatRes will show the totality for central government, a supplement to local systems Only showing data that exist already, but in a structured way Including only central-government spending Not including general transfers (pensions etc) Mainly expenditure, not cost Assumption Most countries in the same situation: - large amounts of relevant public-sector performance data not easily available to the public Strong efforts needed to make such data available to a larger audience – increasing the use and the impact of the data

Central government sectors and units Central government sectors and units StatRes Citizens, voters Politicians Public sector employees Scientists, students Media, organisations USERS STATISTICS AND INDICATORS Display meaningful information about input, output, outcome Main challenge: Select indicators and secure data quality

Current scope: for discussion StatRes has initial focus on uncomplicated sectors, for showroom effect At an early stage show avaliable results and facts, rather than wait for a "perfect" system StatRes will display results "as is" to trigger - new sectors - new indicators - better quality - quest for relations (between inputs and results) Many useful indicators are not dependent on national accounts

National accounts development in Norway vs. StatRes BACK-UP Some observations Relatively high quality in national accounts COFOG adapted in national accounts Strong regulations resulting from cooperation/agreement with EU Some challenges Outsourcing makes it hard to determine input (resource) mix Partial public financing og certain services difficult to treat How to measure real costs, and not only expenditure - include depreciation etc. Eliminating correctly for transactions within public sector (tracing) Considering creation of satellite accounts for StatRes use

System framework: Making all of central government displaying results according to a common framework Input Activity / unit Output Outcome Other determinants Efficiency Effectiveness Other measures: - Prioritisation, coverage, quality 1) Displaying data for all agencies according to this framework will be a major step towards public-sector transparency No ambition to estimate economic value of the outputs 2) Objectives 1) Many indicators may fall into several possible groups/classifications 2) Unlike the national accounts, wich estimates prices for outputs

Introducing performance based financing in the health sector Cost efficiency 2) Technical efficiency 1) 1) Per manyear 2) Per NOK Average number of days on waiting list Number of days on wating list reduced… …however no signs of increased cost efficiency Source: Norsk Pasientregister BACK-UP

COFOG General public services Executive and legislative organs, financial and fiscal affairs, external affairs Foreign economic aid General services Basic research R&D General public services General public services n.e.c Public debt transactions General transfers within government Defence Military defence Civil defence Foreign military aid R&D Defence Defence n.e.c Public order and safety Police services Fire-protection services Law courts Prisons R&D Public order and safety Public order and safety n.e.c Economic affairs General econ., comm. & labour affairs Agriculture, forestry, fishing, hunting Fuel and energy Mining, manufacturing and construction Transport Communication Other industries R&D Economic affairs Economic affairs n.e.c Environmental protection Waste management Waste water management Pollution abatement Protection of biodiversity and landscape R&D Environmental protection Environmental protection n.e.c Housing and community amenities Housing development Community development Water supply Street lighting R&D Housing and community amenities Housing and community amenities n.e.c Health Medical products, appliances & equipm Outpatient services Hospital services Public health services R&D Health Health n.e.c Recreation, culture and religion Recreational and sporting services Cultural services Broadcasting and publishing services Religious & other community services R&D Recreation, culture and religion Recreation, culture and religion n.e.c Education Pre-primary and primary education Secondary education Post-secondary non-tertiary education Tertiary education Education not definable by level Subsidiary services to education R&D Education Education n.e.c Social protection Sickness and disability Old age Survivors Family and children Unemployment Housing Social exclusion n.e.c R&D Social protection Social protection n.e.c.10.9 BACK-UP

Norway has three administrative levels, relatively decentralised Limited outsourcing; most tasks performed by public sector - E.g. few independent schools Ongoing redesign of regional level (counties) Decentral desicions and responsibilities (municipalities) Limited outsourcing; most tasks performed by public sector - E.g. few independent schools Ongoing redesign of regional level (counties) Decentral desicions and responsibilities (municipalities) Public sector consumption, 22% of GDP Municipalities: 40% Counties: 7% Central govt.: 53% Defence Public order and safety Economic affairs Health Recreation, culture, religion Education (higher) Environmental protection Education (sec. elementary) General public services Housing and community amenities Education (primary school) Social protection Recreation, culture, religion EUR billionMain tasks Key features Source: Statistics Norway