1 / 27

DETA Research Forum 20 June 2008 Evidence Based Policy: the challenges, hopes and possibilities

DETA Research Forum 20 June 2008 Evidence Based Policy: the challenges, hopes and possibilities. Peter Crossman Government Statistician Queensland Treasury. Framework. Advance the evidence base Policy generative research Economic, social and population modelling

kamala
Download Presentation

DETA Research Forum 20 June 2008 Evidence Based Policy: the challenges, hopes and possibilities

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript


  1. DETA Research Forum20 June 2008Evidence Based Policy: the challenges, hopes and possibilities Peter Crossman Government Statistician Queensland Treasury

  2. Framework • Advance the evidence base • Policy generative research • Economic, social and population modelling • Research and policy capacity building • Information dissemination

  3. Economic growth • Critical for economic and social progress and security • Drivers of growth in per capita state income are: • Labour productivity • Labour utilisation, including • Employment, less unemployment • Participation, more activity • Intensity, work longer • Terms of trade • Economic growth accounting decomposition

  4. Chart 1.2 Decomposition of growth in GSI per head, 1993-94 to 2005-06 Source: ABS (Australian State and National Accounts and Labour Force), and OESR estimates

  5. What is productivity? • Labour productivity is output per unit of labour input • Growth in labour productivity has two sources: • Multifactor productivity (MFP) i.e. working smarter, more innovatively, doing things differently and better; and • Capital deepening i.e. improving the use of capital by workers e.g. using new machines, or using machines better. • Problem is – productivity is a treadmill

  6. Innovation • A process of continuous change • Innovation is the essence of productivity – in a sense, it is the same thing • An essential pre-condition to “good” change is information • Evidence based policy and decision making – in government, households and business • Keynes “ When I get new information, I change my mind. What do you do?”

  7. Statistics • An important “type” of information • The “market’ for statistics – supply and demand • Gaps – how would one know? • Priorities – wants, aka demand. • But what about supply? Issues include: • Collection costs – direct and indirect • By-products – administrative, ICT systems • Inertia and lags in recognition, reaction • Confusion and lack of “connection”

  8. Administrative statistics • This is what we already have, and it is growing: • Imperfect understanding of what there is out there • Custodians know, sometimes in isolation • Many potential users do not know and cannot know • Imperfect incentives to inform and change • Many restrictions – some genuine, others unnecessary

  9. More on administrative statistics • There can be no innovation when there are no incentives to change and too many rigidities binding change • Solution? • Provide: • incentives • Remove: • restrictions

  10. More on administrative statistics • Provide incentives? • Use and value of data is the primary incentive • Recognition of this is rapidly increasing • Performance measurement – both strategic and tactical • Indigenous reform – information critical to success • COAG – new SPP arrangements depend on performance measures against targets

  11. More on administrative statistics • Remove restrictions? • Provide legal protection (see licensing below) • Clarify ownership, and roles and responsibilities (the Crown owns it all – agencies are custodial, accountable and statutory officers are ultimately responsible, and professional officers are data custodians, who administer and mange the data for the accountable and statutory officers) • Add transparency (see below)

  12. A note on restrictions • Some restrictions are valid – these should obviously not be over-turned • This includes privacy protection. People and businesses are entitled to privacy, and there are good societal reasons for enforcing privacy protection. • There are some other good reasons for restricting data access, but not many.

  13. More on administrative statistics • We should make information about administrative datasets clear to all potential users • Custodians cannot know of potential uses by users – let the users know that administrative data exist, what its quality is like, and what the access arrangements are • Before we collect statistics directly, we should see what is already collected, what is already there

  14. Transparency • Seeing what is there is the concept of transparency • Minimises costs e.g. search • Creates pressure for improvements in quality and quantity (“if we added an identifier for Indigenous status, this would enable population data to be used for… and would prevent the need for, say, another costly, difficult and intrusive survey…”) • Maximises potential use from innovation

  15. In practice, what can be done? • Need practical actions • Simple, cheap, good returns, early runs on the board, low hanging fruit, take the points • Three practical things to do: • Metadata registries; • Licensing; and • Do not do daft things e.g. lock up all data, commoditise all data, or centralise data.

  16. Metadata registries • Example is the Register of Strategic Information in OESR: • has small proportion of population • centralised • not a good example of what needs to happen • What needs to happen? • All data need to be managed with metadata by data custodians • The metadata should be registered in only one place viz the agency registry • Agency registries must be searchable by the world (remember, metadata ≠ data) • All metadata must be transparent

  17. More on metadata registries • Best generic example • Library catalogues – where there are lots of libraries, in different physical and logical locations, of various types of content, each with a searchable comprehensive catalogue, maintained using professional standards, with the books behind the doors in the stacks.

  18. More on metadata registries • Keep it simple, especially in technology, let statistical professionals (the data custodians) organise it • Do NOT centralise metadata (like RoSI) or data (this is risky, costly and unnecessary) • Make it a core responsibility of custodians and accountable officers • Stick metadata on the agency website (WCMS) and make it searchable by a web search engine (e.g. Google Statistics) • Keeps risks allocated properly, costs down, and minimises errors (e.g. the dangerous practice of there being multiple (sometimes illegal, often out-of-date) copies of some strategic datasets – we need a single clear point of truth)

  19. Licensing • Metadata registries must contain metadata, but not necessarily data • Metadata should clarify access conditions • Data are often currently restricted because many custodians do not have simple, legally valid access to licensing advice or means • Creative commons helps in the great majority of cases – solves this problem • Some data will require more detailed licensing

  20. More on licensing • Queensland approach includes GILF (Government Information Licensing Framework) • Went to Cabinet as an information paper this year • Information on GILF on OESR and QSIC websites • Attracting significant interest in other jurisdictions • Based on creative commons – about 85 per cent of government information can use basic CC licenses • Provides legal certainty

  21. Back to productivity • Clearing the public sector’s information “market” will help MFP, through better information allowing smarter decisions and policy • What about capital deepening? • This is important too • Workers need more and better capital • They need skilling to operate this capital as well as access to it • Information is capital

  22. Information is capital • We readily understand physical capital – roads, buildings, computers, staplers • We pay a lot of lip service to intangible capital • This includes information or, in my world, statistics • Need to go beyond glib phrases and make it happen i.e. properly value information content, and manage it like an asset

  23. But it is not ICT • ICT hardware is commonly understood to be capital – as a “machine” it is understood to add value to a worker • Even software is getting there • But information content is still largely forgotten in the stock of assets • Proof? We are here today discussing what needs to be done

  24. Help the workers • Give the workers – researchers, policy analysts and makers, decision makers – the tools for the job • This includes the actual information content they require for business decisions and for evidence based policy – this includes information which is already held by the public sector • Innovative (but secure) data matching and integration will be enabled and will deliver even more capital • Note that workers in their role as consumers also could use this information

  25. What can we do? • Reduce search costs • Increase volumes – the range of possible uses is unpredictable as it is in the hands of innovators (and we do expect innovation) so expose all metadata • Reduce risks for respondents, users, custodians and accountable officers • Add to our stock of statistical (and other information) capital and improve quality

  26. Productivity and innovation • You cannot command it • You can set conditions for it to happen • The two main “framework conditions” from economic policy principles: • Improve flexibility e.g. remove all unnecessary restrictions – enable change • Provide appropriate incentives e.g. encourage or provide reasons for change

  27. The result • Real improvements in our evidence based capability for policy and decision making is needed: • Real challenges in the economy – ageing, climate, energy, skills • Economic growth comes from productivity • Also societal improvement, with better outcomes • Nothing can be guaranteed, but if we can do this, it would certainly seem to improve, statistically speaking, the likelihood of success!

More Related