|innovation in government
At least compared with other countries, Australia’s public sector is relatively innovative. Looking back a century or more, we have introduced some important governmental innovations before other countries, from votes for women to the secret ballot. More recently, since the early 1980s we have been a world leader in economic policy reform which has included the reform of government economic institutions. Thus, we have corporatised and privatised government owned businesses to improve their productivity and introduced access regimes to enable better commercial access to national monopoly assets, whether they are owned by government or privately. We have designed markets in the place of centrally planned service delivery, as in the case of the Job Network, and produced award-winning innovations for rationalising government service delivery, as in the case of Centrelink.
One important theme of government innovation in Australia has been the way in which we have expanded the use of existing government delivery platforms. Thus we have used our tax administration to deliver the world’s first income contingent loans system in the form of the Higher Education Contribution Charge (HECS) introduced in 1989 and expanded since then.
The Child Support Agency was similarly built around the infrastructure provided by the tax system. Replacing a ramshackle court-based enforcement system, the Child Support Agency enforced maintenance collection through the Australian Taxation Office using administrative formula rather than judicial process to determine the level of maintenance. The efficiency of enforcement was dramatically improved, thus lowering costs and improving the equity of Australian society.
There are other areas in which HECS style loans may be appropriate.
Recommendation 10.1: Consider extending the platform created to enforce payments and administer income contingent loans through the tax system; for instance, by extending income contingent loans for tertiary education outside universities and for sole trader entreprenuers seeking to fund innovative projects.
Innovations such as those outlined above are introduced by those at the top, with ministers announcing policies and senior public servants directing their agencies to implement them. As issues grow more complex, however, those at the top of organisations do not have information that those further down the hierarchy may have.
We have known since before suggestion boxes were first introduced into factories that innovation both within firms and elsewhere is often sparked by middle managers and those at the ‘coal face’ with specific knowledge of the issues. As Borins reports:
The data from both the US and Commonwealth innovation awards showed that frontline staff and middle managers are the most frequent initiators of public management innovations. This is a surprising result, given the traditional impediments to innovations emanating from that level in the public sector. It leads to the question of how public sector organizations can be made more supportive of such innovations.1
Australian governments have been good at encouraging ‘bottom up’ innovation where they have established markets in the place of bureaucratic controls – as for example in the case of the Job Network or emissions trading. Outreach programs like Landcare have also fostered innovation at the community level to solve local environmental problems.
Targets and benchmarks have also assisted innovation in some cases – for example, where they have been imposed in Victoria and South Australia to reduce the compliance burden of regulation. CPI-X (price cap regulation) for utilities has likewise been adopted to maximise utilities’ incentives to innovate. On the other hand, too strong an emphasis on targets and top-down command can in fact be highly inimical to successful innovation and better performance on the front line2.
A critical means by which innovation can be encouraged from the bottom up is to build institutions which actively seek to elicit, encourage and respond constructively to feedback from whoever has a good idea. In this regard the private sector has so far outshone the public sector, and governments around the world have only recently begun to address the issue. So far Australia has not been a leader.
From the late 1970s on, and particularly following the leadership of Japan’s automotive industry, businesses seeking bottom-up innovation have instituted a whole range of strategies to encourage and empower people throughout their organisations to innovate. Ideas for improvement from employees, suppliers and customers have been assiduously cultivated. In addition, business has used awards to provide recognition to individual innovators or successful teams. ‘Suggestion boxes’ have been supplemented with additional resources and privileges for bringing innovative proposals to the ‘front of mind’ of the relevant agency.
One promising institution is Shell’s ‘Gamechanger Program’ which invites employees, suppliers and indeed members of the public to make proposals to improve its operations via a website. Proposals are assessed by a small team that reports directly to the CEO and has access to 10 percent of Shell’s R&D budget. This provides a check on the ability of middle management to block the development of new ideas. Those proposing ideas that are assessed as promising are invited to meet relevant company officials within two weeks of their submission. In 2000, the Singapore Government introduced something similar. See Box 4.
Box 4: Singapore’s Enterprise Challenge
The Singapore Government’s Enterprise Challenge (TEC) initiative assesses and funds innovative and risky proposals that have the potential to create new value or significant improvements to the delivery of public service. Projects are chosen by a board that includes business entrepreneurs, area experts, and senior public servants.
TEC provides an open platform (through funding and test-beds) for radical innovations from any sector that need to be trial tested so that their feasibility and practicality can be proven. Anyone with a promising idea can submit a proposal. 'Public agencies often prefer to stick to tried-and-tested solutions,' said Dr Tan Kim Siew, Chairman of TEC. 'They are hesitant to adopt new ideas or innovations which may fail … thereby wasting their time and resources. Being public officials, they will be concerned with auditors faulting them for wasting public funds. This is where TEC comes in. We take the risk, wholly or partially, for them.'
Source: Singapore Prime Minister’s Office, Public Service Division, 2000.
We know of no such program in the Australian government but Victoria has just begun experimenting with mechanisms for initiating ‘bottom up’ policy innovation with a pilot of a staff based suggestions program which has been enthusiastically received within the Department of Premier & Cabinet. See Box .5
Box 5: Policy Idol in the Victorian Public Service
Policy Idol was a competition run in 2007 to capture staff-led innovation across Policy and Cabinet Group (PCG) within the Department of Premier and Cabinet (DPC). It arose out of a planning day for one of the branches, in which staff expressed the view that they would add value to the organisation and public policy more generally by capturing bottom-up ideas for public sector improvement.
Once the idea was endorsed by PCG, policy officers at all levels were asked to email their policy ideas to a coordinating branch. Entrants were offered the opportunity to further investigate their topics during a week in which they would be taken offline from their normal work tasks.
A total of eight entries were received, in areas including real estate, health care, service delivery frameworks, behaviour change and infrastructure development.
All entrants then ‘pitched’ their idea to a DPC senior executive panel, comprising a Deputy Secretary and an Executive Director of PCG. The prize for the winner was one week ‘off-line’ to conduct a feasibility study into the topic. It is now anticipated that Policy Idol will be run annually and expanded across other areas within DPC.
Source: Victorian Department of Premier & Cabinet.
Mechanisms of challenge
Often a new idea fits uneasily into an old world. When Australia led the world in developing vehicle security technology in the 1990s one might argue that the keyless car was the next logical step. Logical it may have been, but legal it was not. Australian Design Rule 25 prevented the introduction of the keyless car, requiring as it did door locks with four mechanical barrels. Clearly mechanisms must be developed to facilitate innovation in such circumstances. In regulation Australia pioneered a mechanism called 'request and response' which invited feedback on regulation. However the mechanism was not well publicised or promoted, and offered little more than an undertaking to consider improvements. After a period of some years the availability mechanism was terminated, having been used no more than four times in any one year.3
Mechanisms of challenge can be made to address such problems. Thus for instance part of the UK’s attempt to improve regulation involves a public ‘suggestion box’ site4 launched in May 2007 to which anyone can make submissions publicly documenting problems with red tape and proposing means of improving them. Proposals are publicly responded to within 90 days and action on progress is publicly reported. The mechanism has already made some progress in cutting red tape.
The UK has also pioneered other legal innovations none of which have been hugely transformative, but which have attempted to build mechanisms of challenge, and mechanisms by which innovations can take place notwithstanding regulatory and other administrative obstacles.
Building on this, exciting new possibilities are now emerging for government from the collaborative use of internet technologies and platforms otherwise known as ‘Web 2.0’.
In the US, crime is being tracked by neighbourhood, combining city crime statistics with Google’s online maps5 . In Los Angeles, Neighborhood Knowledge California6 identifies communities at economic risk by tracking tax delinquency, fire violations and other signs of deterioration. The federal government has launched several wikis, which permit officials to post information and expand on it until a consensus is reached.
Among the most interesting of these experiments is Intellipedia. Born from an internal essay competition, Intellipedia is a cross-departmental platform of collaborative tools including wikis. Officials at the Central Intelligence Agency (CIA), Federal Bureau of Investigation (FBI), National Security Agency (NSA) and other US intelligence agencies share information and even rate one another for accuracy in password-protected wikis, some ‘top secret’. Users are told, ‘We want your knowledge, not your agency seal’.
The possibilities here are so substantial, so full of promise (and sometimes so challenging to existing cultures) that it is neither possible nor desirable for this report to spell out comprehensively what might or should occur. Furthermore, many developments will and should occur ‘organically’ within agencies rather than in response to the dictates of others.
Be that as it may, there are some principles according to which we may be able to make progress. The most fundamental principle is that Governments should be as open as possible to experiments with Web 2.0 approaches. Importantly, they should seek to learn from those that are successful but should expect, and educate the public to expect, that many initiatives will not fully succeed.
Recommendation 10.2: An advisory committee of web 2.0 practitioners should be established to propose and help steer governments as they experiment with Web 2.0 technologies and ideas. At least five substantial experiments should be established in different areas within two years to be evaluated within three. The Minister for Finance and Deregulation should have carriage of the initiative.
Experimentation and evaluation: a new ‘biological’ model of government innovation
A key theme of modern innovation is the loosening of hierarchical routine. Yet the keystone of such an approach is a culture of evaluation. Without knowing what is working and why, suggestions for improvement and the experiments that emanate from them stand for nothing. Accordingly, it is crucial that evaluation becomes a central part of the way governments operate.
Continual internal measurement of results and evaluation is crucial to the productivity improvements that total quality management has unleashed in business. It has yet to take firm hold within government. And the measurement of performance so that external evaluation is possible is the sine qua non of successful and cost effective government innovation.
Where new programs are brought into existence, one of the most important things to establish at the outset is the ‘base case’ or the state of affairs before the program has had a chance to be implemented. The Panel has been dismayed at how frequently programs are instigated without a coherent evaluation strategy or budget or even baseline data which might be used to determine the success or otherwise of the program after the event.
Box 6: When schemes are capped
Assistance schemes often operate under expenditure caps placing their administrators in a quandary as to how to ration access to a scheme. Here notions of fairness can play havoc with efficiency. Modulating the extent of assistance after the event is a means that is often used – for instance in the Automotive Competitiveness and Investment Scheme. Modulation must be one of the worst possible options available. This is because, once firms are uncertain of the level of assistance they will receive when committing to a project they are unlikely to take that assistance into account in their decision making. What assistance is ultimately received is then a windfall.
Wherever a scheme is capped, projects should be chosen for funding according to scheme objectives, rather than modulation of assistance rates. Further, where such caps are administered, this provides an excellent arena in which to run ongoing, usually small scale, randomised trials – so that hypotheses about what works can be tested in an ongoing way and criteria for allocating assistance optimised.
One major source of recent innovation in government has been massively increased attention to early childhood development. This is largely the result of the quality of the data on the effects of early childhood intervention generated in randomised trials in the US in the 1960s. Economists like Nobel Prize winner James Heckman have been able to use data from the trials to claim with much more confidence than they otherwise would have had, that early childhood intervention has a high rate of social and economic return.
Randomised trials provide a 'gold standard' of data for evaluation and should be used far more in Australian policy making. There is any number of areas in which they would be useful. As Andrew Leigh argued to the Panel:7
[W]hat is needed in tackling our toughest social problems is not more ideology, but an evidence-based attempt to find out what works and what does not.
The Panel strongly agrees. Leigh’s support of randomised trials was echoed in submissions by the Victorian and Western Australian Governments.
As we saw at the outset of this section, some innovations can be implemented satisfactorily from the top down. This will be the case where the innovation is some clear idea and it can be adapted to organisational infrastructure we already have, or which can be readily developed. Furthermore such innovation will typically be capable of being implemented for all citizens or firms – perhaps with a pilot program or two. This fits well with traditional notions of fairness in government services and assistance. Most people instinctively feel the importance of governments treating like people alike.
But in many other cases, innovators will need to tentatively grope towards a solution, trying new things to see if they work out. Some innovations will be genuine experiments, because we cannot be sure how they will turn out, whereas others will nevertheless require a good deal of learning by doing.
The Panel agrees with the Prime Minister’s emphasis on policy trials,8 and his emphasis on government innovation.9 Experimentation is how a great deal of innovation happens particularly in the private sector. Instead of a mechanical model where a specific, well understood innovation is rolled out on a mass scale, some firm will take the initiative and try to make something work (thereby taking the very considerable risk that it will not). If governments really are to seize the opportunities that lie before them to innovate, they must engineer a transition in which this kind of experimental learning comes to be seen as both normal and healthy.
In this context it is important that those working in public agencies are able to run planned experiments in doing things better, and it is equally important that those experiments be evaluated in a robust way so that we improve our knowledge of what works and what does not.10 Mulgan and Albury have proposed a circular, iterative model of innovation in the public sector encapsulated in the following diagram.
Figure 17: How innovators learn as they go
Figure seventeen shows a circular iterative model of innovation which has "Incubating and Prototyping" at the twelve o'clock position, linked to "Replication and Scaling-up" at the 3 o'clock position, linked in turn to "Analysing and Learning" at the 6 o'clock position, linked in turn to "Generating Possibilities" at the 9 o'clock position which is linked in turn back to "Incubating and Prototyping", which also has a two way link to Analysing and Learning" at the 6 o'clock position.
Source: Mulgan, G. & Albury, D. (2003), Innovation in the Public Sector, UK Cabinet Office.
In this world, in addition to their normal work, and in addition to 'top down' driven innovation, there would be active encouragement to people and to agencies to do things in new ways where there was some confidence that they may lead to improvements in service quality or productivity. In addition to an encouragement of low level incremental improvements, a substantial number of experimental and developmental initiatives both within agencies and across them would be explored. A subset of those innovations would be proceeded with after initial evaluation by peers and others. They would be implemented in trial forms and evaluated for their success or failure. Where they worked, learning from the innovations would be propagated for others to learn from and/or scaled up as appropriate.
It will be necessary for the kinds of changes that must take place to be spearheaded by a dedicated institution.
Recommendation 10.3: An Advocate for Government Innovation should be established to promote innovation in the public sector.
The Advocate would:
• operate a scheme similar to Singapore’s Enterprise Challenge;
• provide a source for funds and expertise for conducting randomised policy trials;
• manage a process by which agencies within government, and also firms outside it, were able to challenge established practices, administrative arrangements, or regulation which obstructs beneficial innovation;
• provide specific ‘project facilitation’ assistance to firms seeking regulatory approvals in order to introduce worthwhile and innovative business practices;
• promote networks (including within federal and state and territory governments) to maximise the dissemination of knowledge about worthwhile new approaches to issues faced by public agencies;
• operate as a repository of knowledge and resources to the Australian Government and participating State and Territory government agencies to promote tendering practices designed to maximise the scope for innovation in the supply of goods and services to government;
• establish a high profile national awards system to provide national awards for individuals, and public agencies at the Federal, State and Territory or local government level, that make the greatest contribution to public sector innovation; and
• act in concert with an appropriate university partner such as the Australia and New Zealand School of Government, to hold an annual international conference on innovation in government, with the aim of it becoming the premier international conference on the subject – the Davos of public sector innovation.
Recommendation 10.4: A rigorous policy of evaluating all Australian Government innovation programs – and other relevant programs – be established. In a way analogous to the requirement that new regulation cannot be implemented without adequate regulatory impact analysis, a policy should be adopted whereby new programs cannot be implemented without an adequate evaluation strategy and funding for evaluation including the collection of ‘base data’ to evaluate the effects of the program.
Effective evaluation and ongoing policy learning is fundamental to effective governance of innovation policy and programs, and is further covered in Chapter 12, 'governance of the innovation system'.
Leveraging the federal system for innovation
This evaluation role can cross inter-governmental levels. Despite its frustrations and inefficiencies there is much to commend the federal system. Current work in COAG is focused on minimising the downsides of federalism – by seeking to harmonise state regimes where that makes sense to minimise the transactions costs of operating across state lines.
On the other hand, the structure of the Commonwealth has placed limits on the use of this technique. One major objective of the current COAG agenda is to untangle some of the prescriptiveness with which the Australian Government controls the States and Territories with special purpose payments under S96 of the Constitution.
The central advantage of federation is that of pluralism. Not only can different preferences be met and possibilities realised depending on local values and economic and social circumstances, but there’s also the chance to enhance innovation through diversity, experimentation, identification of what works and dissemination of learning from the outcomes of our experiments. Indeed in the United States, some federal politicians promote policies specifically designed to encourage experimentation amongst the states with a view to working towards a national solution in an incremental and experimental way.11
As pluralism is one of the key characteristics that make a federal system work, many have argued that we need to improve connections and collaborations in international and national contexts. In particular, we need to establish a more collaborative environment for the different levels of government to jointly pursue policy and program development and, where appropriate, the joint and seamless delivery of programs and services. It is essential for innovation and breaking down 'silos'.
As the COAG reform council has emphasised, often the problem with innovation and reform in government is one of resources.12 For this reason, interstate trade in public policy expert services should be encouraged, with States and Territories being rewarded for becoming centres of policy excellence in different areas. Following independent evaluation, where a State or Territory is identified as having developed leading expertise in some area of public policy, the Commonwealth should encourage the purchase of expert policy advisory services from that State by other States with a view to disseminating best practice.
To play to its strengths, one important theme of the federation should be to encourage policy and delivery innovation in the various States and Territories. The current rethinking of the financial underpinnings of federalism offers the perfect opportunity to commence work on this important agenda.
Recommendation 10.5: Experimentation in innovative policy and administration should be a major theme of the current refashioning of federal relations. States and Territories should be able to bid for federal funds to pioneer innovative approaches and to have their innovations properly and independently evaluated. This could be taken up within the COAG National Partnership Rewards payments currently being negotiated.
Innovation in procurement
In Australia, the criteria for government purchasing is often highly specific – with the service or product identified and specified in its entirety – and also cost driven – with a demand to obtain the service or product at the lowest cost. However, because government procurement contracts can be large relative to the market, its purchasing power can stimulate innovation so long as the tendering process allows it.
First, if the supplier is providing an innovation as part of its service to the government, that innovation and its development is likely to benefit local industry. For this reason, unlike the private sector, the government should internalise the local spillover when evaluating competing tender offers. As Professor Joshua Gans has argued:
[G]overnment should take an active role in putting innovative elements into its procurement practices. This not just a short-term role but a leverage of the government’s existing role. This could be achieved by adding a ‘novelty’ component (to promote new solutions to a share of government service outsourcing) and a ‘spillover’ component (to add explicit consideration of the benefit a government contract could yield on the provision of new innovations throughout the broader economy).13
In this way, government can act as a catalyst for innovation by paying a premium for it.
Second, numerous submissions indicated a prescriptive government response on procurement with little scope for innovation. This appears to be driven by an extreme aversion to bearing risk. The Warren Centre submission (p. 10) quotes a response received during research into public sector clients’ attitudes: 'innovation increases risk and should be avoided at all costs unless it cannot be avoided because of otherwise unachievable timing or cost objectives.' They go on to raise concerns about senior procurement officers who are 'not properly equipped with the technical or commercial knowledge of how to buy and use engineering services; the result can be a heavy dependence on legalistic tendering and contractual frameworks ... that simply do not and cannot take advantage of the … engineering innovation available' (p. 9). Further, the CEO of Hatrix, an Australian Medical Software Services SME, notes 'the almost total reluctance of those in charge of government procurement processes to risk their reputation on a small Australian company if there is a large multinational in the mix'. It is instructive that Hatrix had to go to New Zealand to get its first customer.
The Australian Government should be open to changing its tendering processes to allow experimentation and innovation. As the UK Government Report Transforming Government Procurement commented:
It is much easier to evaluate the costs and benefits of a tried and tested product, rather than something that may not have previously been used in practice, or may not even exist at the time the Government first considers using procurement as a means of solving a complex delivery problem. However, if a new and better solution is already developed or could be made available, this might provide better value for money than a tried and tested product.14
Smart and innovative government purchasing involves purchasing in such a way that the door is left open for novel approaches to be pursued where these would improve outcomes and/or lower costs.
In the UK, this more innovative approach to procurement is being implemented. For example, in prison mattress purchases and disposal the UK government used advanced purchase commitments to solicit 35 innovative solutions (including different materials and alternative recycling systems). Thus, beyond simple cost requirements, zero-waste mattress solution options were considered including a full ‘cradle to cradle’ supply management solution. This allowed innovation in environmental as well as economic management.
In a similar vein, but with lower-powered incentives, the US Small Business Innovation Research (SBIR) scheme, established in 1982, requires US Government research agencies with budgets of over $100 million for external R&D to spend 2.5 percent of that budget assisting small business to engage in R&D that has the potential for commercialisation and public benefit.
The SBIR is structured in three phases.
Phase I, which typically involves SBIR assistance of $100,000 or less, determines the technical merit, feasibility and potential for commercialisation of the proposed R&D efforts and the quality of performance of the SME prior to providing Phase II support.
Phase II, which involves SBIR assistance of less than $750,000, continues R&D efforts with funding based on the success of Phase I and further assessment of scientific and technical merit and commercial potential.
Phase III involves the SME pursuing the commercialisation objectives resulting from the Phase I/II R&D activities with non-SBIR funds. In some Federal agencies, Phase III may involve agency supported follow-on non-SBIR funded R&D or production contracts for products, processes or services intended for use by the US Government.
The scheme has run for a sufficient time to be reviewed extensively and the reviews have been positive. The SBIR scheme is regarded by many as defining best practice in the field and has been widely imitated in countries such as Finland, Sweden, Russia, India, the Netherlands, Korea, Japan, Taiwan and the UK.15
Australia adopts an SBIR type approach in Defence procurement that is well regarded. The recent statements by the Parliamentary Secretary for Defence Procurement (reported in Australian Financial Review, 25 June 2008) are encouraging in regard to the target of 20 percent of the annual Defence procurement budget in Australia to go to Australian SMEs (up from 15 percent), but will only be effective in facilitating innovation if risk is shared.
The Panel considers that introducing a much broader and better-funded SBIR type scheme would fill an important niche in the Australian innovation system and promote a problem-solving and learning culture within the public sector.
Recommendation 10.6: The Australian Government should recognise its role as an active participant in facilitating innovation through procurement practices. In this context, the Government should:
actively manage its ability to enable and demand innovation in procured services and products, given its significant presence as a major purchaser;
in procurement, be open to participating in risk sharing in relation to innovation demanded;
explore the use of forward purchase commitments as a means of fostering more innovative approaches to government procurement; and
work with the States and Territories to implement a pilot Small Business Innovation Contracting program based on the US SBIR design principles, to strengthen the growth of highly innovative firms in Australia.
The Advocate for Government Innovation should operate as a source of expertise and seed funding for the resourcing of such approaches to procurement.