MLD 102 2014 class 6

MLD 102: Class 6, September 22 2014
When contracting-out and the like works
(and when it does not)
More lessons about using NPM tools to ‘get things done’
Matt Andrews
How did we get to a case on concessions in
Mexico in the early 90s?
• Our focus is on how to ‘get things done’
– Getting governments etc. to implement policies, deliver services, etc. more
often, more efficiently, and more purposefully
• Many thought about this challenge in the first half of the 2oth century,
but solutions offered tools that did not always deliver
– Classical administrative and bureaucratic theory
• Which still has a huge influence today
• But was criticized in the west in the 70s and 80 (It seems these tools don’t always work)
• At around the same time that developing country governments were emerging
• New ways of thinking emerged in the west, around something that
became known as ‘new public management’
– Flatter and decentralized government, using performance metrics,
outsourcing, privatization and contracting out
• A series of ideas that influenced many developing countries and offered solutions to the
problem of ‘getting things done’
• But are these tools the panacea or are they also subject to conditions?
The questions matter
• Places like Messaria seriosuly consider these tools:
– Using targets for health care providers
– Contracting out parts of the health system, decentralize (or keep
decentralized other parts)
• You will need to think about these tools in your strategies for
getting things done
– SYPA, PAE, and beyond
• What do you need to know to make sure you choose them at
the right time and with the right concerns in mind?
– We discussed this w.r.t performance information on Wednesday
– Today we look at concessions
What happened in Cancun?
• What was the problem?
• Why did they choose a concessions arrangement?
• Was it a success?
• Why and what should be learned from this experience?
Lessons in using contracting out, etc. to get things done, 1
• This is an important tool (Nichols):
– “Over the last several decades, in governments at all levels
throughout the world, the public sector’s role has increasingly
evolved from direct service provider to that of an indirect provider
or broker of services; governments are relying far more on networks
of public, private and nonprofit organizations to deliver services.”
• Potentially offers a way to:
– Raise money, save money, improve efficiency, deliver more quickly
• But it should not be seen a panacea (Nichols):
– “Outsourcing is by no means a perfect solution. Some agencies don’t
have the metrics in place to prove in advance that outsourcing a
service will save money. Problems from poorly conceived contracts
can create cost increases that surpass the costs of in-house services,
and if there’s shoddy contract oversight, a government is vulnerable
to corruption and profiteering. The privatization of public services
can erode accountability and transparency, and drive governments
deeper into debt.”
Lessons in using contracting out, etc. to get things done, 2
• You need basics to use the tool (Nichols) in the USA, beyond:
– “The states most successful … created a permanent, centralized entity to manage and
oversee the operation, from project analysis and vendor selection to contracting and
procurement. For governments that forgo due diligence, choose ill-equipped contractors
and fail to monitor progress, however, outsourcing deals can turn into costly disasters.”
• And the problems need to be addressable by outsiders
– Something which is often not the case when the problems are about existing public
sector dysfunction, politics, etc. AND solutions are unknown.
– “…outsourcing deals are really about risk…You’re taking the risk of the unknown and
dumping that on your supplier…You’re outsourcing a problem to a company that has
limited control over the root cause of the problem.” The only way for a public-private
partnership to work, he suggests, is to drive transformation from within the agency. And
that’s the hard part. Red tape usually prevents governments from making significant
modifications, and private companies lack the authority to enforce real changes.”
• So: Choose the right tasks and have the necessary systems in
place; and ensure that the contextual challenges (politics etc.)
won’t inhibit an ‘external agent’
Lessons in using contracting out, etc. to get things done, 3
• Learning from NPM experience generally:
• Schick: The New Zealand model (late 80s early 90s) of ‘Government by Contract’
– “New Zealand has gone to extraordinary lengths to create conditions under which formal contracts
are negotiated and enforced”
Created entities to lead, accrual system to finance it, a capital charge system to capture full costs, etc.
• Schick: Some key lessons emerged:
– Better suited to “matters that can be specified in contracts, such as the purchase of outputs”
– “Robust contracting depends on voluntary, self-interested action. Sometimes, however, self-interest
defeats the government’s collective interest.”
– Contractualism may weaken traditional values of public service, personal responsibility, and
professionalism (a checklist approach to accountability).
– Contract-like arrangements do not create arms-length relationships in the public sector, nor do they
enable the government to toughen its insistence on performance. (Exit options?)
– Govt. officials attribute most of the improvement in government performance to the discretion given
to managers rather than to formal contracts. (Do contracts make the difference?)
– Contracting is not costless. Negotiating, enforcing contracts entails enormous transaction costs.
• Schick: Developing countries are not a good fit for this…
– The demands of these tools make it ‘out of reach’
– The informality of the systems in many developing countries make it really risky
Decreased accountability etc.
– Countries should rather create reliable systems of controls etc. (Weberian bureuacracies?)
Lessons in using contracting out, etc. to get things done, 4
• Manning says the results of NPM in development are limited:
– “The effect has been mixed at best with some improvements in efficiency
and mixed effects on equity. On the downside…the transaction costs of
radical reforms tend to outweigh the efficiency gains of unbundling, and
that reforms that pursue the separation of purchasers from providers may
have decreased accountability — and that, consequently, inequity has
• Partly because needed public expectations are muted:
– “From the government side, the sound of any nascent consumer discontent
other than at the local level is drowned out by the far louder noise of donor
conditionalities. The New Public Expectations in many developing countries
seem likely to remain modest, and sidelined by the realities of aid.”
• And because the public sectors lack necessary formality:
– “It is an empirical observation that predictable resourcing, credible
regulation of staff and credible policy are prerequisites for effective
contract-like arrangements.”
So, should we never use these tools in development?
• Ehasn and Naz: it may not solve problems or be
– “Critics argue that the argument in favour of introduction of NPM system is not
convincing, particularly as it has failed to address the crucial issues of ethics,
accountability, non-partisan distribution and administration…It has also been
argued that the very institutional and organizational structure of developing
countries poses stupendous problems to successful implementation of the
reform programmes.”
• Da Cunha Rezende: the change process itself may not
succeed, given extant interests in governments
• But Dan and Pollitt say it can be useful:
– “An adequate degree of administrative capacity, sustained reform over time and a
‘fitting context’ are the main factors which can tip the scale for the success of these
management instruments.”
• But how do we think about these factors?
Idea about ‘fit’
Notice different strategies for different situations
Manning: Legacy of NPM in developing countries
Pathways to change
incapable and
• high state
• high
• little
• non-participatory
achieved through
locally initiated
changes to create
capable local
incapable but
achieved through
locally initiated
changes to create
movement through
some combination
of local voice and
strengthened local
capable but
• high state
• medium
• basic
• non-participatory
• medium state
• high
• little
• participatory
movement achieved
through whole
of government
reforms, including
budget transparency
and financial
capable and
achieved by
economic interests
and raising public
expectations of
service delivery
• medium state
• medium
• administrative
• participatory
movement towards
efficiency and
service quality
achieved through
NPM — but also
through traditional
Note: This figure has benefited from discussions with Brian Levy and Navin Girishankar in
So: we need to think about fit to context
• Polidano…these reforms have fostered success and failures.
– Where they fit, they help
– Where they don’t fit, they hurt
• We need a contingency approach to deciding if they fit:
– “There are at least two lessons in contingency here. First, political and
indeed administrative leadership makes a big difference in ensuring
that reforms overcome the implementation hurdle (the output stage).
Second, to secure the desired outcome— in our particular case, better
organisational performance as a result of improved staffing—reform
initiatives have to be adapted to prevailing local circumstances.”
• This approach should be the one we use for our entire toolbox:
– “These conclusions, one might object, are not exactly striking in their
originality. True. But this is precisely the point. The success or failure of
new public management initiatives depends on the same fundamental
determinants identified by researchers in relation to previous
generations of reforms.”
We have a whole lot of
classical administration,
NPM and
tools in
our toolbox
How do we
Work out what to use,
When and how…
Next session
• We will move towards a strategy for fitting
the tools to the context in the next weeks
– Including a strategy to build political support
and an appetite for change
• A first question in every context is simply:
“What does ‘getting things done’ mean?”
• We will discuss on Wednesday…
– Two cases of revenue agency reform

similar documents