In recent years governments have realised the power of guilting people into fulfilling their duties not because the rules say they have to, but because failing to do so would let others down.
Telling citizens most people pay their tax on time makes them more likely to do so themselves. Tell them how many of their neighbours pay on time and the compliance rate jumps again.
And it turns out social pressure works on public service bosses, too.[pullquote] “There are now 40,000 fewer working age people receiving welfare payments than three years ago…” [/pullquote]
Although it might not seem fair, New Zealand has discovered that when it comes to tackling long-standing social disadvantage that cuts across agency responsibilities — an area of public policy traditionally made difficult by muddled lines of responsibility and barriers to coordination — holding public service executives collectively responsible for improvements is more effective than focusing on individual performance.
Experience with NZ’s Better Public Service Results program demonstrates that appointing a lead individual to each of five groups set up to pursue clearly articulated social outcome improvements put too much emphasis on the person in charge, resulting in weaker feelings of commitment by other team members.
A shift to collective responsibility — including the awarding of bonuses based on collective effort — in theory made freeloading easier, but “seems to produce the best outcomes”, pushing staff to ensure the group achieves something of value.
This is the interesting conclusion drawn by UNSW’s Rodney Scott and Ross Boyd of the State Services Commission of New Zealand in a new paper for the IBM Centre for the Business of Government discussing the success factors behind the Better Public Service Results approach.
The Better Public Service Results program
Responsibility-sharing is just one of the many factors the researchers believe have made the experiment a success.
The BPS Results program involves publicly committing to 10 priorities, which incorporate specific targets across five groupings: reducing long-term welfare dependence, supporting vulnerable children, boosting skills and employment, reducing crime and improving interaction with government.
It ran for five years (2012-2017) and the final status report was published in March.
Although some targets were not reached, there were dramatic improvements for all 10 results. The number of infants not receiving vaccinations fell by two-thirds, for example. Other problems were cut in half, such as the number of children not enrolled in early childhood education and the number suffering from rheumatic fever.
There are now 40,000 fewer working age people receiving welfare payments than three years ago, thanks to more intensive and individualised case management and bureaucrats actively developing partnerships with local businesses.
Following the success of the first BPS Results program, the government has committed to launching a new set of 10 results and targets in 2017. Work is currently underway to identify what these results and targets will be.
So why did New Zealand’s push work where others have failed? Scott and Boyd’s paper finds 13 practice insights that made the program a success.
Choose targets carefully
Focus on a few problems. Articulating just 10 clearly defined targets gave agencies the chance to spend time and money achieving them, lessening the chances of trying to do too many things at once. Being selective meant government chose problems that were most amenable to being resolved in this way — ones that had persisted despite many attempts to fix them and merited trying out a new approach.
Only choosing a few goals means you have to work out what’s most important. NZ’s targets were chosen for their broad impact — either directly (reducing the number of people on long-term welfare) or indirectly (reducing the number of children developing rheumatic fever, an acute respiratory disease most commonly occurring in childhood, which required making improvements to housing and health services that would benefit a wide range of disadvantaged New Zealanders).
Involve other agencies in selecting problems to be addressed. The problems selected were ultimately chosen by cabinet, but after lengthy consultation with departments. Agencies felt more committed to achieving a target they have been involved in selecting, argue Scott and Boyd. Additionally, agencies were well-placed to provide technical advice on what problems were important and what targets would be achievable.
Build on existing relationships when selecting results to pursue. Collaboration was more successful when the parties involved had developed trusting relationships through succeeding together on smaller practical projects. Trust reduces transaction costs associated with monitoring performance, and agencies are more likely to commit their own time and resources if they have confidence that their partners will do the same.
Measure intermediate outcomes. Designing the program around results raises questions about what should form the basis of the targets. Governments often monitor outputs, which are easier to determine but do not necessarily show real world improvements, though they are commonly urged to focus on outcomes.
But while ultimate outcomes are the most useful indicator, they can take many years — even decades — to materialise. Intermediate outcomes trade the benefits of outcome measures, which are intrinsically valuable to society, with output measures, which respond rapidly and predictably to changes and allow government to adapt its approach. Although the ultimate aim of increasing infant immunisation rates is to reduce preventable disease, it is much easier and faster to observe changes in the former.
New Zealand’s experience suggests that adaptive management was most meaningful when the impacts of new actions could be observed within six months. This also helped motivate public servants, demonstrating the improvements their own work was making to citizens’ lives.
Align results, targets and measures. Failing to specify the magnitude of the improvement to be pursued for each target can mean any progress is seen as achieving the goal, so quantifiable targets can help create a sense of urgency and ambition in public servants.
One of the targets, “New Zealand businesses have a one-stop online shop for all government advice and support they need to run and grow their business”, suffered from a lack of clarity. The target was redefined as: “Businesses’ costs from dealing with government will reduce by 25%”. Another measure was developed focusing on perceived effort required by businesses. “However, the misalignment among the three metrics caused confusion and delay, as public servants debated whether they were responsible for implementing a proposed solution (a one-stop online shop), reducing cost, or reducing required effort,” the authors note.
Commit publicly. Although governments tend to be vague about exact goals, lest citizens and the media focus more on the failure to hit targets than improvements made, doing so signalled to public servants that the targets would not be quietly retired when the minister moved onto something else. Happily, local media were more likely to frame data updates positively, focusing on what progress was being made even if targets were not reached.
Hold leaders collectively responsible. New Zealand’s trial and error experience with three different approaches to accountability — individual responsibility for outcomes, individual responsibility for behaviours, and collective responsibility — showed the latter to work best. The State Services Commission is currently exploring whether there are limitations to this technique. The authors also note that the normative importance of the results was at least as significant a motivator, too.
Get started and learn by doing. Central agencies played a role in assisting responsible agencies to overcome the barriers they faced. Previous attempts at cross-agency work showed agencies found it difficult to get started — groups were reluctant to make decisions until they were working well together, but they couldn’t work well together until they were willing to undertake decisions.
While central agencies generally allowed groups to develop their own ways of working, one requirement was that chief executives would prepare and submit an initial action plan for how they would first address the problem. This helped kick-start the process and to ensure that effort began without delay.
Start simply. Where trusting relationships did not already exist, a partial solution to building trust developed over time. Some cases took the time to first practice working together in simpler arrangements — such as information sharing, cooperation, and coordination — to help build mutual trust before attempting more complex collaboration. These tended to be more successful than those that leapt straight to the most involved and interconnected solutions. Achieving quick wins was also important to secure group commitment to solving the problem, particularly for newly formed groups, and reinforced the need to collaborate.
Limit group size. Although some believe it’s vital to include all affected stakeholders, the most successful cases limited core decision-making to two or three agencies, utilising tiers of involvement when necessary. In some cases, more than three agencies had an interest in the outcome, but greater progress was made when two or three critical partners formed a core group and involved others on an as-needed basis for information sharing and coordination. As group size increases, the transaction costs of coordinating the group increase, while the responsibility felt by each agency decreases.
Signal shared responsibility. Public servants, who tend to work in hierarchies, may look for signals that demonstrate who is really in charge in a collaborative context — and who will be responsible for success or failure. The most successful groups carefully orchestrated equal commitment at all levels through cascading governance groups.
Reporting on Progress
BPS Results involved lots of regular public reporting. Two documents were released every six months — the ‘dashboard’ (three pages) and the ‘snapshot’ (one page). Both include a colour-coded progress assessment and a line graph, with the dashboard also including explanatory text. A mix of qualitative assessments (such as colour-coded progress indicators) and quantitative indicators struck a balance between making results easy to understand and providing transparency about what exactly is happening for those more sceptical.
Report on trends. Focusing on trends, rather than static numbers, allows for contextualisation of progress. It demonstrates the improvements that have been made, giving more information than just whether or not a target has been hit.
Share success stories. Central agencies in New Zealand deliberately engaged in positive reinforcement of departments’ work by celebrating success, so reporting was generally well received by public servants, bolstering motivation and innovation.
Beginning in 2013, agencies began to submit short text or video descriptions, showing innovations that had a positive impact. The most notable briefs from the previous six months were included in the dashboards and advice to cabinet. More detailed case studies were shared through the State Services Commission website.