Mapping Public Sector Innovation Units in Australia and New Zealand - 2018 Survey Report
←
→
Page content transcription
If your browser does not render page correctly, please read the page content below
Mapping
Public Sector
Innovation Units
in Australia and
New Zealand
2018 Survey Report
Michael McGann, Jenny M Lewis, and Emma Blomkamp
This survey was funded by a research grant from The Australia and New
Zealand School of Government (ANZSOG)
Cite this publication as: McGann, M, Lewis, JM and Blomkamp, E (2018)
Mapping public sector innovation units in Australia and New Zealand: 2018
survey report. Melbourne: The Policy Lab, The University of Melbourne.
The Policy Lab at the University of Melbourne addresses contemporary
challenges of politics and policy-making through research centred on three
main themes: opinion and expertise in politics and policy, policy systems, and
projects that deal with specific issues and policy sectors. Related to these
areas, The Policy Lab explores three core questions:
What shapes policy decisions?
How do policy systems and processes operate?
What works in policy design?
If you are interested in The Policy Lab's projects or partnerships, you can
contact the team at the-policy-lab@unimelb.edu.au
Twitter: @ThePolicyLab
Medium: www.medium.com/the-policy-lab
Web: arts.unimelb.edu.au/the-policy-lab
The Policy Lab
School of Social and Political Sciences
University of MelbourneCONTENTS
Executive summary / 4
Introduction / 7
About the Survey / 9
Organisational size and history / 11
Background and skill-sets of staff / 15
PSI units’ relationship to government / 18
Policy areas and innovation domains / 31
Methods of PSI Units / 29
Conclusion / 33
References / 34
Appendix A: List of participating PSI units by location / 35
Appendix B: Principal component analysis of innovation domains
and methods / 36
3EXECUTIVE SUMMARY
Public sector innovation (PSI) units are Over half were established within the past
increasingly being established and two years.
commissioned by governments to bring new
While we found several examples of very large
insights and approaches to policy design and
PSI units in Australia—including some with
the delivery of public services. According to
more than a hundred staff—the survey results
previous estimates, worldwide, there are
overall highlight the extent to which PSI units in
approximately 100 such units based within
Australia and New Zealand are very small
governments alone while new units are being
organisations:
created at ‘a rate of at least one a month’ (Price
2015). This report, based on a survey of PSI About half of PSI units employ five staff or
units in Australia and New Zealand undertaken fewer
in February 2018, suggests that the actual 12 of the PSI units employ at most two
number of PSI units worldwide may be people.
significantly higher than previously thought.
As a result, PSI units may have to frequently
Although we surveyed PSI units based both
draw in external expertise or second staff from
within and outside of government, we
other agencies and departments to carry out
identified at least 26 PSI units based in various
work. This is implied by the extent to which we
levels of government in Australia and New
found PSI units in Australia—both within and
Zealand alone. There are a similar number of
outside of government—were utilising
non-government units and mixed-organisation
consultants and/or consultants:
types regularly undertaking public sector
innovation work with, or on behalf of, In the six months prior to the survey,
governments in Australia and New Zealand. This government-based PSI units each hired just
includes organisations such as The Australian under four consultants or contractors on
Centre for Social Innovation (TACSI) and the average while units based outside
Behavioural Insights Teams, many of which government (independent PSI units) hired
collaborative extensively with state over six consultants or contractors on
governments. And this is only counting the PSI average to carry out work on their behalf
units that responded to the invitation to
Government-based PSI units seconded an
participate in the survey.
average of just under three staff from other
All this suggests a vibrant public sector government agencies and departments in
innovation landscape is emerging in Australia the six months prior to the survey
and New Zealand. This is reflected in how
More consultants/contractors than
recently many of the PSI units surveyed were
employees carried out work for
established:
independent PSI units in the six months
Six of the government-based PSI units have prior to the survey, suggesting that there
been operating for 12-months or less may be a supply chain of public sector
4innovation in Australia and New Zealand such as ‘Graphic design’ and ‘Service or
whereby governments commission user experience design’
independent PSI units who, in turn, further
Government-based PSI units tend to recruit
subcontract work.
primarily from within the public sector
The relationship of PSI units to governments in
While these PSI units tend to recruit at least
Australia and New Zealand is highly varied,
some staff from outside government,
although the survey results suggest that most
including community-sector organisations
units can be classified as either government-
and design agencies, only about a third
controlled PSI units (based within and funded
report recruiting ‘many’ of their staff from
by government) or government-enabled units
outside government.
(non-government units that rely extensively on
government funding). Unlike some prominent Similarly, the range of policy sectors and public
international examples such as the UK service delivery areas that PSI units work across
Government’s Policy Lab, very few (if any) is wide:
government-based PSI units in Australia and
More than half of PSI units undertake work
New Zealand appear to charge out their
in the area of ‘Social issues, housing and
services despite regularly undertaking work for
welfare’
other agencies and departments:
‘Public administration and governance’,
Most of the government-PSI units surveyed
‘Education’, ‘Health’, and ‘Indigenous and
were funded by the national or federal
Maori issues’ are other prominent policy
government, and based within a parent
sectors that many PSI units work on
department or agency
Only two were co-owned by multiple 16 PSI units in Australia and New Zealand
agencies or departments, and both of these work on ‘Transport’ policy
were NZ-based PSI units 12 PSI units undertake work on ‘Policing,
10 of the Australian PSI units surveyed were crime, and the justice system’, including
based within a state government eight government-based PSI units.
department or agency Within these policy sectors, we have identified
The survey results also indicate that three distinct domains of innovation that PSI
independent PSI units regular carry out units are involved in:
work for state government departments
and agencies. 1. Policy development and reform: involving
identifying or scoping problems, consulting
The skills-sets, qualifications and capabilities of with stakeholders, scaling and spreading
people working within PSI units in Australia and new approaches, supporting and
New Zealand are very diverse: developing partnerships, developing policy
Qualifications in ‘Sociology or social work’ proposals and reforms, and working on
and ‘Management and business systemic change;
administration’ are the most common, 2. Evaluation and systems improvement:
Many PSI units are staffed by people with based around evaluating
formal qualifications in design disciples programs/trials/pilots, incorporating
technology into public administration,
5organisational change management, and and New Zealand, which converge around three
business systems or process improvement; main frameworks:
and
1. A Human-Centred Design framework: this
3. User and customer-experience:
is associated with the use of ‘interviews
understanding users’ experiences,
and/or empathy conversations’; ‘focus
generating ideas, piloting/prototyping
groups’; ‘ethnographic methods’;
solutions, and service or customer
‘citizen/stakeholder engagement through
experience (re)design.
workshops, walkthroughs, and other
collaborative approaches’; ‘user
Many PSI units undertake work in all three
testing/prototyping’; and ‘systems thinking
domains of innovation, although activities in the
or mapping’;
domains of User and customer-experience and
Policy development and reform appear to be 2. An Evidence-based framework: this is
the predominant focus of PSI units, with associated with the use of ‘RCTs’;
individual PSI units appearing to frequently ‘Behavioural Insights’; ‘Survey research’;
undertake activities in both these domains. ‘Research/evidence reviews’; and the
‘Analysis of existing (big) data sets’; and
They also bring a variety of methodological
3. An Agile methods framework: this is
frameworks and approaches to the work that
associated with the use of ‘design sprints’;
they are undertaking in these domains and
‘agile or lean project management’; and
policy areas, but especially a suite of methods
‘challenge prizes, awards, and open
associated with Human-Centred Design. This
innovation programs’.
underlines how the spread of PSI labs
internationally has been associated with Of these three, HCD was the methodological
growing interest in the application of ‘design framework most frequently employed by the
thinking’ to policy. This relationship between PSI units surveyed, and was associated with PSI
the emergence of PSI units and ‘design thinking’ units undertaking activities in the domains of
approaches to policy and public sector reform is Policy development and reform and User and
largely confirmed by the survey results on the customer experience.
methods that PSI units are using in Australia
6INTRODUCTION
The number of public sector innovation (PSI) The reasons why governments are turning to
units – sometimes also referred to as these ‘islands of experimentation’ (Tõnurist et
government innovation labs or i-teams - has al. 2017) are varied. Some commentators argue
spread rapidly in recent years, both that the proliferation of PSI units reflects a
internationally as well as in Australia and New heightened emphasis on evidence-based
Zealand. Prominent international examples policymaking by government departments and
include the Danish Government’s MindLab, and public agencies.
the UK Policy Lab. MindLab is a cross-
Others have associated PSI units with the
departmental innovation unit, jointly owned by
pursuit of ‘open government’ agendas and
the Danish Ministries of Industry, Employment,
initiatives to promote transparency,
and Education, while Policy Lab is formally
accountability, and the empowerment of
located within the Cabinet Office of the UK
citizens through new technology and data
Government although it works across the whole
sharing platforms (Gryszkiewicz et al. 2016, 7).
public service.
This seems to be a particularly important focus
PSI units thus have varying relationships to of the work of PSI units in Latin American and
government and differences in their relative Caribbean countries (Acevedo and Dassen
proximity to executive decision-making. This is 2016). Many others identify the spread of PSI
what Geoff Mulgan (2014) describes as the units with the recent upsurge in interest in co-
‘radical’s dilemma’: working at a distance may design and the application of various ‘design
enable PSI units to develop more frame- thinking’ practices to public policy (Bailey and
breaking alternatives to the status quo but at Lloyd 2016; Bason 2013; Fuller and Lochard
the risk of being ignored and marginalised; 2016; Mintrom and Luetjens 2016).
while working more centrally within
On occasion, commentators have indeed
government may enable units to more directly
defined PSI units by their emphasis on using
influence policymaking but at the risk of being
innovative design methods to ‘reform and
co-opted and shifted from radical to
change the way government operates’ (Bason
incremental change.
and Schneider 2014, 35) and ‘involve all
Importantly, not all PSI units are based within stakeholders in the design process’ (Fuller and
government. Many operate instead as Lochard 2016, 1). This reflects the view among
independent organisations that work with—and many practitioners and commentators that the
for—governments on a commissioned basis to previous ways in which governments have
support agencies and departments looking to sought to design and implement policies are no
innovate in policy design or public service longer suited to the complexity of today’s policy
delivery. This includes some of the most and social challenges (Kimbell 2016); that the
frequently cited examples of PSI units, such as systemic nature of policy challenges in areas
Nesta’s Innovation Lab and The Australian such as health, social services, employment and
Centre for Social Innovation (TACSI) (see, for education (among others) leaves governments
example, Puttick et al. 2014; Selloni and with little choice ‘but to innovate’ (Puttick
Staszowski 2013). 2014, 20)—principally through embracing a
more ‘experiment-oriented’ (Fuller and Lochard
72016, 14) approach that, in turn, requires supported by a grant from The Australia and
drawing on capabilities and skills-sets usually New Zealand School of Government, and feeds
not available in the public sector (Carstensen into wider ongoing research that The Policy Lab
and Bason 2012, 5). is carrying out on the sector throughout 2018.
This includes detailed case studies of five PSI
While the age of the innovation lab has arrived,
units to further explore their collaborative
little is known about what these labs do, how
governance arrangements and the challenges
they differ from other public sector change
and opportunities they experience in
agents and policy actors, and what range of
contributing to policy innovation and reform at
approaches they are bringing to contemporary
different jurisdictional levels. Further details
policy challenges (McGann et al 2018). Further,
about this ongoing work are available on the
although several international mapping reports
project web page:
and practice guides on PSI units have been
http://go.unimelb.edu.au/ix86.
produced within the past five years, the
emerging landscape of PSI units in Australia and
New Zealand has yet to fully documented or
mapped. Interested parties had been
wondering:
How many PSI units are now operating in
Australia and New Zealand?
How recently have they been established
and what are the key challenges they face
to their operation and survival?
What are the key levels of government and
policy areas that they are working within?
What are the distinct capabilities and
approaches that these units are drawing
on and bringing to innovation in public
service delivery and policy design?
To address these questions about the
emergence of PSI units in Australia and New
Zealand (NZ), The Policy Lab at the University of
Melbourne undertook an exploratory survey of
the sector in early 2018. The survey was
8ABOUT THE SURVEY
The survey was carried out online from 29 This involved directly approaching PSI units
January to 25 February 2018. For the purposes within and outside of government that we were
of the survey, PSI units were defined as any unit already aware of to participate in the survey,
or team that was ‘established for the purposes publicising the survey via The Policy Lab’s
of supporting public or social innovation’ website, social media and mailing list, and via
including both ‘units within government, or the contacts within government and the wider
public sector, as well as non-government public sector. For example, an email about the
organisations and labs that work with survey was circulated to ‘Heads of Labs’ within
governments on public sector innovation.’ the Australian Public Service via the Public
Sector Innovation Network (PSIN), which is
The question of how to define PSI units is a
managed by BizLab within the Department of
contentious issue due to the multiplicity of
Industry, Innovation, and Science. The PSIN also
organisations that feature in discussions of the
included an article about the survey in its
emergence of innovation units and labs, and
weekly email circular, ‘Bits of News’, which goes
their heterogenous nature (McGann et al. 2018;
out to more than 3,600 subscribers, including
Tõnurist et al. 2017). For this reason, and
people working within Local and State
reflecting the exploratory nature of the survey,
Government and non-government
we opted for a definition that enabled us to
organisations who are interested in public
include or exclude potential participants based
sector innovation.
upon whether they self-identified as being
established to support public sector innovation. Additionally, the survey was promoted to
Subsequent questions probed participants members of the Australia and New Zealand
about the extent to which they are funded by School of Government – which includes public
governments and undertake projects for servants working within the New Zealand
government departments and public agencies Government, the Australian Government, and
at various levels. The answers that participants all State governments in Australia – and via
gave to these questions indicated that the non- Twitter using the hashtag #psilabs. This is
government units and teams that responded to recognised as a commonly followed hashtag by
the survey nonetheless regularly worked for, practitioners within the field of public sector
and with, government partners on public innovation, particularly among those involved in
service and policy design innovation projects. innovation units and teams (Williamson 2015).
Finally, individual survey participants were
As the total number of PSI units operating in
asked to nominate other PSI units or teams that
Australia and New Zealand is unknown, we
they were aware of who might wish to take
adopted a multifaceted and snowball sampling
part.
approach to recruiting potential participants.
9A total of 52 PSI units and teams took part in lab in exchange for receiving personal
the survey, once duplicate and unusable office/studio space from Moreland City Council.
responses had been removed from the data A full list of participating units and teams is
file. This included 13 responses from NZ-based provided in Appendix A, with 11 out of the 52
units and 39 responses from PSI units and respondents indicating that they have offices in
teams based in Australia. As Table 1 shows, 26 multiple locations.
out of the 52 units and teams that responded
While the survey included both government-
to the survey were based within government –
based PSI units as well as units operating as
albeit at different levels – while a further 23
non-government organisations – and also units
units were independent from government in
innovating in public service design as well as
the sense that they had operational
teams working on policy innovation – the
independence and were not subject to direct
number of units that responded to the survey
oversight by a government department or
suggest that the total number of PSI units
agency. Three PSI units were mixed-
worldwide has been substantially under-
organisations that operated as a partnership
estimated in previous research. For example, it
between government and a community-sector
is commonly reported that ‘Worldwide there
or non-profit organisation. In New Zealand, this
are over 100 Policy Labs, approximately 65 of
included the Tamaki Mental Health and
these are in Europe’ (Whicher 2017). The term
Wellbeing Team, which is a partnership
‘policy lab’, as used in these estimates, is more
between the Auckland District Health Board, its
narrowly defined than how the term ‘PSI unit’
primary care partners, and the Tamaki locality.
has been used to determine eligibility for
In Australia, this included Designing Out Crime,
participation in this survey. Nevertheless, the
which is a research partnership between the
survey has identified that there are at least 26
NSW Department of Justice and the University
government-based PSI units in Australia and
of Technology Sydney; and the Moreland Civics
New Zealand alone—units that closely
Lab, which is an experimental lab for local
resemble the government innovation and policy
government issues in which designers, artists
labs featured in international reports.
and researchers contribute to the work of the
Table 1: Profile of Participants
Based Independent Mixed Total
within from govt organisations
govt
New Zealand 5 7 1 13
ACT 10 2 12
NSW 4 5 1 10
QLD 1 1
SA 1 2 3
VIC 4 6 1 11
WA 1 1 2
Total 26 23 3 52
10ORGANISATIONAL SIZE AND HISTORY
Previous studies have characterised PSI units as Only four PSI units reported that their
‘islands of experimentation’ (Schuurman and establishment was an initiative of an elected
Tõnurist 2017, 9) within the public sector, official or member of government. This
emphasising their fluid structure, relatively suggests that the emergence of PSI units within
short life span, and small size in comparison to government in Australia and New Zealand is
other public sector organisations (Tõnurist et al. being driven by public managers and
2017). This is reflected in our survey findings, as administrators, rather than by politicians or
reported in Figure 1. Almost a quarter of the PSI elected officials seeking to promote a policy
units based within government (6 out of 26) agenda of government innovation and public
have been in existence for 12 months or less, sector reform. This marks one way in which the
and three quarters (20 out of the 26) have been recent proliferation of PSI units can be
established within the past three years. distinguished from earlier public reform
agendas and discourses, such as the
By comparison, the non-government or
‘government reinvention’ labs that were
independent PSI units tend to have a more
established in hundreds of US Government
established organisational history with almost
departments and agencies in the early to mid-
two thirds (15 out of 23) indicating that they
1990s under the direction of the Clinton
had been operating in their present form for
administration (Thompson and Ingraham 1996).
three years or more. When asked about how
the decision to establish their unit or team The results reported in Figures 2 and 3 show
came about, the overwhelming majority of that PSI units—both within and outside of
government-based PSI units (70.8%) reported government—are overwhelmingly very small
that it was ‘an initiative of the head or chief organisations in terms of the number of staff
executive officer of a government/council they employ. Although seven PSI units reported
department or agency.’ employing more than 20 staff, approximately
half of the units surveyed directly employed five
staff or fewer.
Figure 1: How long have labs existed?
INDEPENDENT (23)
GOV-BASED (26)
0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100%
· 12 months or less · More than 1 but less than 2 years
· More than 2 but less than 3 years · 3 to 5 years
· More than 5 but less than 10 years · 10 years or more
11Indeed, almost a quarter employed Zealand had seconded 70 of its staff to
considerably fewer staff than this with one in government departments and public-sector
five government-based units reporting agencies. In contrast to seconded staff, the
employing no more than two direct employees survey results indicate that PSI units make
and one in four independent units reporting the frequent use of consultants and that this is a
same. consistent pattern across units based within
and outside of government. For example,
The small size of PSI units suggests that they
across government-based PSI units the ratio of
may rely heavily on either working with external
consultants (used within the previous six
consultants or seconding staff from (other)
months) to employees was just under half (i.e.
departments and agencies when carrying out
units had commissioned one consultant for
projects. However, we found that few PSI units
nearly every two of their direct employees)
relied on seconding staff from government
while among independent PSI units it was
departments and agencies. Only 10 of the
higher still (1.6 consultants per direct
government-based PSI units surveyed reported
employee).
having seconded ANY staff to work with them
within the previous six months. And although
the mean number of seconded staff reported Lab Example 1: Co-design and Innovation Lab
by government-based PSI units was 2.9 people, Based within Tauranga City Council, New Zealand.
the magnitude of this number was largely
Established within the past two years.
accounted for by two PSI units who each
reported seconding a total of 30 or more staff Employs three staff members.
from other government departments and Additionally engaged four consultants as well as
agencies (the Project Office within Policy seconding four additional staff from other agencies
Innovation and Projects Division of the and departments to work on projects within the
previous six months.
Australian Government’s Department of Prime
Minister and Cabinet, and the Service
Lab Example 2: BizLab
Innovation Lab within the New Zealand
Government’s Department of Internal Affairs). Based within the Australian Government’s
Department for Industry, Innovation and Science.
PSI units reported being even less likely to place
Established within the past two years.
their own staff into government departments
and agencies to work on projects, with only six Employs 15 staff members.
government-based PSI units reporting having Within the six months prior to the survey, this was
seconded their own staff to work within also supplemented by three interns and two staff
another department or agency within the members seconded from other government agencies
and departments.
previous six months. The comparatively high
mean number of staff that independent PSI One of the few (6 in total) government-based PSI
units that reported seconding its own staff (3 staff
units reported seconding to government
members) to another agency or department to
agencies (5.4 staff) can be partly explained by undertake project work.
the fact that one international design agency
with multiple offices in Australia and New
12Figure 2: How PSI units are staffed (Government-based units)
20+ 2 staff
staff or
fewer
Employees (mean)
Proportion
11 to of units by Consultants used in last 6 months (mean)
20 staff staff size
3 to 5 Seconded FROM a gov. dept. within last 6 months
staff
Unit staff seconded to work within a gov. dept. in last 6 months
6 to 10
staff
Ratio of consultants to employees
Figure 3: How PSI units are staffed (Independent units)
20+ 2 staff
staff or
fewer Employees (mean)
Proportion of
units by staff Consultants used in last 6 months (mean)
11 to size
20 staff
Seconded FROM a gov. dept. within last 6 months
3 to 5
6 to 10 staff Unit staff seconded to work within a gov. dept. in last 6 months
staff
Ratio of consultants to employees
Figure 4: Ratio of consultants to employees by organisation size
Units with 2 or less staff
Units with 3-5 staff
Units with 6-10 staff
Units with 11-20 staff
Units with 20+ staff
13The reliance of PSI units on engaging external
Lab example 4: Co-Design Studio
consultants may necessitate from the small
number of staff they directly employ. A design consultancy firm head-quartered in
Melbourne with a sister office in Brisbane.
Although units with a higher number of direct
employees generally reported commissioning It has been operating for over five years, employing
more consultants and contractors than units 12 staff members.
with fewer direct employees, smaller PSI units This is supplemented by a mix of interns and
nevertheless reported engaging a contracted consultants, with Co-Design Studio
considerable number of consultants in engaging 10 consultants and employing 10 interns in
the six months prior to the survey.
comparison to their organisational size. This is
illustrated by the high ratio of consultants to
direct employees among PSI units with small Example 5: Smart City Office
numbers of employees, reported in Figure 4 Based within the City of Melbourne.
above.
Established within the past three years.
Employs a total of 40 direct staff. Within the six
Lab example 3: ThinkPlace months prior to the survey, this was supplemented
An international strategic design consultancy firm by the engagement of 10 consultants.
that has been operating for over 10 years.
It has offices in Canberra, Sydney, Melbourne,
Wellington, and Auckland, and employs
approximately 80 staff.
Within the six months prior to the survey, it had
seconded 70 of its staff to working within
government departments and agencies.
Conversely, it rarely utilises consultants or staff
seconded from elsewhere, only engaging four
consultants and one staff member from a
government department or agency to work on
projects
14BACKGROUND AND SKILL-SET OF STAFF
Globally, the spread of PSI units has been government-based PSI units. This reflects the
heralded as part of a movement ‘that is bringing participation of several behavioural insights units
knowledge and practices developed in other in the survey, although it is surprising that
fields into the heart of the public service’ (Puttick comparatively few (less than one in three) PSI
et al. 2014, 5). This is thought to be especially units reported having staff with formal
true of PSI units that are committed to applying qualifications in economics – another discipline
‘design thinking’ to policymaking (Bason 2013; that is strongly associated with drawing on
Mintrom and Luetjens 2016). But to what extent behavioural insights for policy. What Figure 5 also
is this true of PSI units that are operating in illustrates is the high proportion of PSI units
Australia and New Zealand? We sought to employing people with formal qualifications in
address this question by surveying units about design disciplines such as ‘graphic design’ and
the formal qualifications held by their staff, and ‘service or user experience design’ (18 PSI units
the sectors from which they have recruited staff in total reported employing staff with these
— for example, whether their employees have qualifications). This is the case across both
considerable private sector experience, or have government-based and independent PSI units,
previously worked in design agencies or in although it is especially true of the latter with just
academia or other research organisations. under half (47.6%) of independent PSI units
indicating that their staff held formal
The findings reported in Figure 5 show that PSI
qualifications in graphic design.
units employ staff with very diverse skill-sets.
‘Sociology or social work’ was the discipline most Figure 5: Skill sets of staff in PSI units
frequently identified by PSI units overall in terms Data science or analytics
of the formal qualifications held by their staff, Political science
with 20 PSI units in total reporting that they had Psychology
staff with such qualifications. This likely also Economics
reflects the types of policy issues that PSI units Marketing and communications
work in (See Figure 18), with ‘Social Issues, Management or business…
Accounting or finance
Welfare and Housing’ the main policy area that
Law
PSI units report working on by some distance.
IT or software development
Among government-based PSI units, the most
Industrial or product design
frequently reported qualification held by staff
Service or user experience design
was in ‘Management or business administration’ Graphic Design
(13 out of 22 government-based PSI units that Engineering
responded to this question) followed by ‘Law’ (11 Architecture or planning
PSI units). Only nine government-based PSI units Education
reporting having staff with formal qualifications Sociology or social work
in ‘Sociology or social work’. Following business Anthropology
and legal qualifications, ‘Psychology’ was the next 0 5 10 15
most commonly reported discipline by Independent Government
15The ‘design’ background of PSI unit staff is ongoing operation. Not surprisingly, ‘risk
further reflected in the proportion of units who aversion’ was the most frequently reported
report that either ‘some’ or ‘many or almost all’ challenge identified by government-based PSI
of their staff worked in a design agency prior to units, with 18 government-based units reporting
joining their unit or team. As Figure 7 shows, that they had encountered this as an obstacle to
almost 70 per cent of independent PSI units (13 their operation or establishment.
out of the 19 units responding to this question)
Figure 6: Sectors from which Government-based PSI unit
reported that some, many or almost all their staff staff have been recruited
had prior experience working in a design agency.
This compared with only eight government-based Overseas innovation units
PSI units that reporting having recruited some or Other innovation units within
your country
many staff from design agencies (See Fig. 6).
Academic or other research
Rather, government-based PSI units appear to be organisations
heavily reliant on recruiting people already within Other private sector
businesses
the public sector, with 18 government-based PSI
Management or consultancy
units reporting that ‘many or almost all’ their firms
staff previously worked in the public sector, and a Design agencies
further five reporting that ‘some of their staff’
Non-government community
previously worked in the public sector before sector
joining their unit or team. Public sector
When government-based PSI units do recruit 0 5 10 15 20 25
from outside the public sector, it appears that
Some Many or almost all
they are marginally more likely to do so from
non-profit organisations rather than commercial
Figure 7: Sectors from which Independent PSI unit staff have
enterprises. What is noticeable across both been recruited
government-based and independent PSI units is
Overseas innovation units
the proportion of units (over a third) that have
Other innovation units within
recruited staff from either an overseas or your country
domestically-based PSI unit. Academic or other research
organisations
Staffing issues, and the capabilities and skill-sets Other private sector
of personnel emerged as important operational businesses
challenges that the PSI units surveyed reported Management or consultancy
firms
(See Figs. 8 and 9). For example, half of PSI units
Design agencies
overall ‘agreed or strongly agreed’ that the
‘difficulty of attracting highly skilled staff’ was an Non-government or
community sector
important challenge they experienced. Among
government-based PSI units, 16 PSI units Public sector
reported that ‘lack of capabilities and skill-sets 0 5 10 15 20
within the organisation’ was a key challenge they
Some Many or almost all
experienced to either their establishment or
16The challenges identified by PSI units in this Bureaucracies function to bring ‘predictability
survey echo long-standing concerns about the and order’ (Puttick et al. 2014, 3) but this can
capacity for innovation within public sector engrain an aversion to risk within their
organisations and whether bureaucratically organisational cultures not least of all because of
structured departments and agencies are suited the political accountability and public scrutiny
to developing ‘radical new solutions’ to complex they are subject to (Schuurman and Tõnurist
social and policy problems (Carstensen and 2017, 7). This takes us to a consideration of PSI
Bason 2012, 3). units’ relationship to government and the
executive, which we examine in the next section.
Figure 8: Challenges reported by Government-based PSI units Figure 9: Challenges reported by Independent PSI units
Lack of operational capacity
Lack of operational capacity
within the organisation
within the organisation
Risk aversion within Risk aversion within the
the organisation organisation
Funding constraints Funding constraints
Lack of commitment to driving innovation of change
from senior internal/external decision makers Lack of commitment to driving
innovation or change from senior
internal/external decision makers
Difficulty of securing endorsement from host org. leadership Difficulty of securing
endorsement from the unit or
team's host organisation's
leadership
Lack of capabilities and skills-
sets within the organisation Lack of capabilities and skills-sets
within the organisation
Difficulty of attracting
highly skilled staff Difficulty of attracting highly
skilled staff
-20 -10 0 10 20 -15 -10 -5 0 5 10 15
Disagree or strongly disagree Agree or Strong Agree
17PSI UNITS’ RELATIONSHIP TO GOVERNMENT
The units and teams participating in this survey clients and community-sector organisations, we
have widely varied relationships to government can similarly map PSI units in Australia and New
and the public sector. Although half of the units Zealand along these dimensions.
surveyed could be described as based within
The results reported in Figures 10 and 11 suggest
government, 23 out of the 52 units that took part
that PSI units in Australia and New Zealand tend
in the survey were independent from government
to operate predominantly as either government-
in the sense that they were not based within a
controlled units – i.e. units based within the public
government department, agency or public sector
sector and wholly-funded by government – or
organisation but operated as independent
government-enabled units – i.e. units located in
organisations that partnered with public sector
non-government or for-profit organisations but
organisations to drive innovation and reform in
which rely on government funding. For example,
policy design and public service delivery.
14 out of the 20 independent PSI units that
Elsewhere we have distinguished four broad
answered this question estimated that either
organisational forms of PSI units based on the
most of all of their annual budget is derived from
extent to which they are funded, and subject to
‘contract funding from government clients’.
oversight, by government (McGann et al. 2018):
However, six of the PSI units based outside of
1. Government-controlled units that are based government reported that they received no
within/owned by a government contract funding from government clients and so,
department(s) or public sector agency and in this sense, could best be described as entirely
wholly funded by government independently-run units. Only one government-
2. Government-led units that are based based PSI unit reported receiving any funding
within/owned by a government from non-government sources. What is also
department(s) or public sector agency but evident from the results reported in Figure 10 is
only partly funded by government that very few government-based PSI units (only
3. Government-enabled units that are based four) operate on a fee-for-service model whereby
within or operate as a non-government other departments and agencies (within the
organisation but rely to a significant extent public sector) engage them to work on projects on
on government funding (mainly through a commissioned basis. This is reflected in the fact
contracts) that 17 of the government-based PSI units
4. Independently-run units that are based within reported that they were entirely funded by ‘direct
the private or third sector and which receive budget allocation’, which has been described
no government funding (these types of elsewhere as a ‘sponsorship model’ (Whicher
innovation units are most analogous to think 2017, 7) for funding innovation units.
tanks that seek to influence public sector
The survey findings clearly suggest that, unlike PSI
innovation and reform through independent
units internationally such as the UK Policy Lab,
research, advocacy, and the promotion of
only a very small fraction of government-based
ideas)
PSI units in Australia or New Zealand operate on a
Drawing on how PSI units responded to questions cost-recovery or commercial model whereby they
about the extent to which they are funded by charge client departments for the work that they
government or rely on funding from private do.
18Figure 10: How Government-based PSI units are funded Figure 11: How Independent PSI units are funded
External fundraising/grants
24 from foundations or 11 8 1
External fundraising/grants philanthropic organisations
from foundations or
philanthropic organisations
Contract funding from private
24 enterprises or businesses
9 11
Contract funding from
private enterprises or
businesses
Contract funding from not-for-
23 1 profit clients
9 9 2
Contract funding from not-
for-profit clients
Contract funding from
20 4 government clients
6 3 11
Contract funding from
government clients
Direct budget allocation from
25 17 government/host 14 4 11
Direct budget allocation department(s)
from government/host
department(s)
None Some of budget None Some of budget
Most of budget All of budget Most of budget All of budget
This is somewhat surprising given that, as the data Focusing on PSI units based within government
reported in Figure 12 shows, a significant and the public sector, we next consider the
proportion of government-based PSI units are different levels of government that they are
undertaking work for other government located in and the extent to which government-
departments and agencies. PSI units based PSI units in Australia and New Zealand are
predominantly report that the projects they cross-agency initiatives or nested within single
undertake are either requested by the parent departments and agencies.
department or agency in which they are based, or
As Figure 13 shows, despite the significant
are self-generated initiatives of their own unit or
number of units that undertake work across
team – indicating that PSI units generally have a
departments, 17 of the government-based units
considerable degree of organisational autonomy
surveyed reported that they were owned by, and
(Tõnurist et al. 2017, 15). Nevertheless, 12 of the
based within, a single government department or
government-based units reported that the
agency.
projects that they work on sometimes or often
originate from another government department
or agency.
19Figure 12: Origin of Government-based PSI units’ projects Figure 13: Location of PSI Units within Government
Another government
department or agency
From the central executive 0 5 10 15 20 25 30
branch (e.g. Department of
Prime Minister and Cabinet Based within a parent government department or agency
or Premier and Cabinet) Co-owned by multiple government departments or agencies
Based within the central executive branch
From within your own unit or
team (e.g. self-generated Operates as a separate public sector organisation
projects)
Your parent department(s) or
agency(ies) (if relevant)
0 10 20 30
0 5 10 15 20 National government (funded by)
Sometimes Quite or very often State government (funded by)
Local or municipal government (funded by)
Only two government-based PSI units reported For the purposes of this analysis, we have
that they were co-owned by multiple grouped government-based PSI units into units
departments or agencies (the MindLab model), based within national government and units
and both were New Zealand PSI units. These based within a state or territory government (we
were: the NZ Government’s Service Innovation have excluded units based within local
Lab, which is located within the Department of government, since only 3 PSI units reported being
Internal Affairs but funded by (and works across) based within local government).
multiple government agencies, and the Auckland
The data suggest that PSI units are very frequently
Co-Design Lab, which is a partnership between
in communication with other government
Auckland Council and eight different central
agencies and departments, particularly on a
government agencies. This funding model was
‘weekly’ or monthly’ basis, although cooperation
unique among the PSI units surveyed, with
across different levels of government is less
Auckland Co-Design Lab being the only unit that
common. In particular, PSI units appear to seldom
reported being funded by or accountable to
engage with local government unless they are
multiple different levels of government – in this
units directly based within local government. On
case central and municipal government.
the other hand, PSI units based within state or
Figure 14 details the survey findings on how territory governments reported frequently
frequently PSI units within government are in engaging with national government, with four
contact (excluding email circulars) with staff from state-based PSI units indicating that they are in
other government agencies and departments, as ‘weekly’ contact with someone from a
well as the extent to which PSI units based within central/federal government department or
national government agencies and departments agency. This is in contrast to the lack of weekly
are interacting with people from state and local communication between units based within
government and vice versa. national government and people from lower
levels of government.
20Figure 14: Frequency of contact across agencies and levels of government
PSI units based in national government PSI units based in state or territory governments
Someone from local Someone from local
5 1 1 2 4 2 1 2
government government
Someone from a Someone from a
state/territory state/territory
1 2 2 3 1 3 2 3
government department government department
or public agency or public agency
Someone from a Someone from a
(another) central or (another) central or
4 5 1 1 2 1 4
Federal government Federal government
department or agency department or agency
Someone from your unit's Someone from your
parent department or 1 1 6 unit's parent department 1 1 4
agency or agency
0% 20% 40% 60% 80% 100% 0% 20% 40% 60% 80% 100%
Never Less than quartery Never Less than quartery
Quarterly Monthly Quarterly Monthly
Weekly Weekly
Note: only units that are based within either national or state
governments are compared, due to the very small number of
units (3) based within the local or municipal government.
The survey also asked PSI units to report on the There was also a high-level of ‘quarterly’ contact
extent to which they are in contact with a range of with PSI units overseas, further pointing to the
external stakeholders (including private sector strong collaborative linkages between PSI units.
businesses, individual citizens, and community What was somewhat surprising, however, was the
sector organisations) and participating in number of government-based PSI units that
networks of public sector innovation units reported ‘never’ communicating with an
nationally and internationally. individual member of the public (6 out of 21), a
representative of a citizen or consumer
The data suggest that networking with other PSI
interest/advocacy group (6), or a representative
units is high among government-based PSI units,
of a community sector organisation. The lack of
with half reporting that they are in at least
engagement with citizens and community
‘monthly’ communication with a representative
stakeholders is particularly surprising in light of
from a public sector innovation network or
the very strong emphasis that PSI units seem to
professional association within their own country.
place on understanding citizens’/users’
experiences, consulting with stakeholders, and
service or customer experience (re)design in
21terms of how they describe the domains of Figure 16: Networking and frequency of contact with
stakeholders (Independent PSI units)
innovation that they frequently work on (see next
section). This raises the question of how PSI units
are consulting and engaging with citizens to An individual member of the
11 5 8
understand their experiences and co-design new public
approaches and services with them, if many never
A representative of a citizen or
actually communicate with individual members of consumer interest/advocacy 2 11 7 4
the public or representatives of citizen interest group
groups.
Researchers from a university
2 5 4 4
or other academic institution
Figure 15: Networking and frequency of contact outside of
government (Government-based PSI units)
A representative of a
community sector 1 2 5 7
organisation or NGO
An individual member of
6 21 6 6 Someone from a private firm
the public
11 4 2 7
or business
A representative of a
citizen or consumer 6 2 3 6 4 Someone from local
interest/advocacy group 1 2 2 5 5
government
Researchers from a Someone from a
university or other 2 2 12 6 state/territory government 2 8
academic institution department or public agency
A representative of a Someone from a (another)
community sector 5 1 4 7 4 central or Federal government 2 2 5 5
organisation or NGO department or agency
A representative of a public
Someone from a private sector innovation network or 3 3 4 4 1
1 3 4 7 7 professional association…
firm or business
Someone from the OECD
A representative of a
Observatory of Public Sector 1 2
public sector innovation
2 4 4 11 1 Innovation
network or professional
association within your… Someone from a public sector
Someone from the OECD innovation unit in another 11 5 7 1
Observatory of Public country
14 5 3
Sector Innovation
0% 20% 40% 60% 80% 100%
Someone from a public Never Less than quarterly Quarterly Monthly Weekly
sector innovation unit in 2 9 9 02
another country
In contrast to government-based PSI units, the
0% 20% 40% 60% 80% 100%
independent units surveyed reported engaging
Never Less than quarterly with citizens and community interest groups on a
Quarterly Monthly
Weekly very frequent basis. Eight out of 15 independent
PSI units reported being in contact with individual
members of the public on a ‘weekly’ basis, with a
further five units communicating with individual
members of the public on a monthly basis. Similar
patterns were evident in relation to contact with
community sector organisations and
22representatives of consumer interest or advocacy questions about which levels of government they
groups. What also emerges from the data had worked for over the previous 12 months. As
reported in Figure 16 is the higher frequency of shown in Figure 17, independent PSI units were
contact that independent PSI units have with more likely to report having worked on projects
people from state government departments and commissioned by departments or agencies within
agencies compared with people based within a state or territory government than to have
departments or agencies of the central or federal worked on projects commissioned by an agency
government. For example, eight Australian or department of the central or federal
independent PSI units reported being in ‘weekly’ government. Indeed, eight of the independent
contact with someone from a state or territory units reported that the projects they work on
government department (this question only originate ‘quite often’ or ‘very often’ from a state
applied to Australian based PSI units since this government department or agency, with a further
level of government does not apply to New five independent PSI units reporting that they
Zealand). This was in comparison to five out of the ‘sometimes’ work on projects originating from
sample of both Australian and New Zealand state government. Although broadly similar
independent PSI units that reported being in numbers of independent PSI units reported
weekly contact with someone from a central or frequently working on projects originating from a
federal government department or agency. Also, central or Commonwealth government
whereas contact between government-based PSI department or agency, these numbers included
units and local government was rare, 10 the New Zealand based PSI units (which were not
independent PSI units reported being in either asked about their cooperation with agencies and
‘weekly’ or ‘monthly’ contact with someone from departments at state government level). The data
local government. reported in Figure 17 also suggest that
independent PSI units almost never work for the
These findings concerning the level of contact
Department of Prime Minister and Cabinet,
between independent PSI units and different
whereas they are more likely to undertake work
levels of government are further supported by
on behalf of a state Department of Premier and
how independent PSI units responded to
Cabinet.
Figure 17: The different levels of government that are commissioning Independent PSI units
Quite or very often 1 8 6 3 8
Sometimes 2 4 8 4 5
0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100%
Federal/National - The Department of Prime Minister and Cabinet
Federal/National - A Federal government department or agency
A local or municipal government
State/Territory - The Department of Premier and Cabinet
State/Territory - A State government department or agency
23POLICY AREAS AND INNOVATION DOMAINS
Internationally, the rise of PSI units has been Most of the units surveyed, however, reported
framed as a response by governments to the working on at least four different policy or public
increasingly ‘complex and systemic’ nature of service delivery areas.
today’s policy and social challenges (Public Policy
Figure 18: Main policy areas that PSI units work on
Forum 2013, 1). As Kieboom (2014, 9) observes of
the context in which PSI units have emerged: ‘The
latest trend in our quest to fix the global
challenges of the 21st century is to “lab” complex Indigenous or Maori issues
issues.’ This raises questions about the principal
policy and service delivery areas that PSI units and Transport
teams work in, and how they are approaching
policy design and public sector innovation within Urban design and planning
those areas.
Public administration and
PSI units undertaking the survey were asked to governance
nominate the five policy or service delivery areas Policing, crime, the justice
that they had worked on the most during the system
previous 12 months. As the data reported in
Health
Figure 18 shows, ‘Social issues, housing, and
welfare’ was by far the main policy area that PSI
Social issues, housing and
units reported working on especially among the welfare
independent PSI units, 16 of which had worked in
this policy area within the previous 12 months. Education
This was followed by ‘Public administration and
governance’, although government-based PSI Employment
units reported working equally on both areas.
‘Health’ and ‘Education’ were other key areas of Economic Development
focus for the PSI units that participated in the
survey along with, unexpectedly, ‘Transport’ and Environment or conservation
‘Policing, crime and the justice system.’ Indeed,
0 5 10 15 20
one in three government-based PSI units reported
that they had worked on ‘Policing, crime, and the Independent Government
justice system’ within the previous 12 months. Six
of the PSI units surveyed (all government-based Another way of conceiving of the focus of PSI
PSI units) reported working only on one policy units’ work is in terms of what might be described
area or issue. These areas were: transport (1); as the domains of innovation they are involved in.
policing, crime, and the justice system (2); social That is, within the policy or service delivery areas
issues, welfare, and housing (1), and health (2). that they are working, at what stages of the policy
or innovation cycle (Figure 19) are PSI units
predominantly operating?
24Figure 19: Comparing the Policy and Innovation Cycles
Opportunities
and challenges
Generating
ideass
Support and
Agenda setting
Maintainence
Develooping
and testing
Evaluation Formation Making the
case
Delivering and
implementing
Implementation Adoption
Growing and
scaling
Changing
Source: Adapted from (Howlett and Ramesh 1995; Puttick systems
2014)
PSI units were asked about the different stages of ‘very’ or ‘quite’ frequently on piloting/prototyping
the policy and innovation cycles they solutions. Conversely, far fewer units reported
predominantly work on and the extent to which that they were involved at the evaluative and
they are actively involved in policy development scaling end of the policy cycle. For example, the
and reform as opposed to innovating on how proportion of PSI units who reported that they
existing services are delivered within established ‘very frequently’ worked on ‘evaluating
policy frameworks. Figures 20 and 21 report the programs/trials/pilots’ was just under 30 per cent.
survey responses on these items for government- The proportion of PSI units who reported that
based and independent PSI units respectively. The they frequently worked on policy development
data suggests that PSI units and teams are very and reform was also low, particularly among
heavily involved at what might be termed the government-based PSI units.
earlier stages of the policy cycle, namely:
Allied with PSI units’ focus on the earlier stages of
identifying/scoping problems; generating ideas;
the policy cycle is their emphasis on stakeholder
and piloting/prototyping solutions. For example,
engagement and exploring citizens’ and users’
over 90 per cent of independent PSI units and well
experiences of public services and programs. This
over half of government-based PSI units reported
is reflected in the very high frequency with which
that their unit or team is ‘very frequently’ involved
PSI units report ‘consulting with stakeholders’ and
in identifying or scoping problems. Almost all PSI
working on ‘enhancing government-citizen or
units reported working either ‘quite’ or ‘very
stakeholder communication/engagement.’ Along
frequently’ on ‘generating ideas’, with nearly 85
with identifying and scoping problems and
per cent of independent PSI units reporting that
generating ideas, engaging with stakeholders,
they ‘very frequently’ worked on ‘generating
citizens and users to understand their experiences
ideas’. Over 60 per cent of the government-based
appear to be key activities of PSI units in Australia
PSI units surveyed, and almost all of the
and New Zealand.
independent PSI units, indicated that they worked
25You can also read
