Management Excellence Award, Wits Business School
5 October 2006
Thank you for this award. I am still not too sure what I have done to
deserve an award for management excellence. I manage just a few people in an
office space not much bigger than this room. I am pretty sure that you have not
spoken to any of the people who occupy that office space because if you did I'm
sure that I would not be the recipient of such an award.
But thanks anyway.
We all find ourselves muttering from time to time that "if you want to get
the job done properly, you have to do it yourself". This is the problem of
agency. There isn't time to do all the budget spreadsheets yourself, fix the
leaking tap, restructure the aging politician's retirement plan, help the kids
understand trigonometry, do your tax return, water the plants and work on
global peace and prosperity all at once.
So for at least some of these projects or part of each of them we rely on
other people.
But here is the problem. We rely on other people but they are seldom
perfectly or completely reliable. The great disillusioning factor which we have
to come to terms with is that we can't do it all ourselves and the power of
delegation always carries the risk of disappointment.
Armed with the apparatus of microeconomics I and a cursory reading of Adam
Smith, of course, you might say specialise! The market will provide! Produce
the commodity in which you have a comparative advantage, purchase everything
else!
An obscure economist named Ronald Coase explained nearly 70 years ago, why
we don't in fact, rely on the price mechanism alone to address all the
limitations of our individual competences. There are transaction costs and they
help explain why some things are organised across markets and other things are
internally arranged. Within the family there are conventional or sometimes
negotiated, divisions of responsibility. Within the firm there is a hierarchy
of duties and a structure of rules, sanctions and rewards. And in the National
Treasury, the Minister says what has to be done.
Or rather that is how Ministers' of Finance fantasise about things.
The idea of transaction costs is an interesting and useful one. It is partly
about incomplete information and over the past 30 years, an extensive
literature has evolved that attempts to explore how we deal with information
shortfalls, how managers enjoy particular kinds of discretion as a result of
imperfect information for example and how improved information standards,
better accounting rules, perhaps can assist in addressing problems of
agency.
I was having a discussion with Tom Boardman the other day. He said that
earlier that day, he'd met with all 450 of his branch managers. There were
people from across the country, a diverse bunch of people with different
capabilities, competencies and approaches. But he could fit all these people
into one hall and have a discussion on their concerns. He told me this story as
he empathised with the Minister of Education who has 27 000 branch managers. We
have 27 000 school principals in our country. We simply cannot get them all
into a single hall or venue and structure a discussion in a sensible fashion.
The transaction costs of engineering such a discussion would be enormous even
with modern information technology (IT) and communications systems. And so we
organise our schools into districts, regions and provinces. Information and
perspectives are lost along the chain.
Agency, incomplete information and the economics of organisations and how
they are governed are areas of economic inquiry that have enjoyed something of
a resurgence in recent years.
We have not made enough progress yet in applying these ideas to the public
sector. And yet problems of agency, incomplete information and misaligned
incentives are surely at the centre of the concept of government failure or
bureaucratic failure, so it is something of a puzzle that public sector
economics has not mobilised these analytical tools more effectively.
It is not that we don't have good illustrative case material.
Consider, for example, an invention that dates back at least a hundred years
and arguably is the single most significant source of power of the State other
than brute military force. I am referring to what is politely known as
"withholding", an arrangement built into the tax code which hugely reduces the
administration and compliance costs of collecting personal income tax from
employees essentially by introducing an "agent" in the form of the employer as
a disinterested intermediary between the receiver and the taxpayer.
Two further inventions have also greatly strengthened the hand of revenue
during the 20th century, the idea of social security which combines a
withholding tax and a promise of future benefit entitlements and the
self-enforcing invoice-based value added tax, which operates through the
opposing interests of buyers and sellers of intermediate products.
These are important features of the public finances of the modern state but
their underlying logic is not part of the standard toolkit. Yet this underlying
logic is essentially economic in nature, it is about the structure of
incentives between principal and agent and how transaction costs can be reduced
when these are well-ordered.
Which is an elaborate way of talking about some pretty simple and hugely
important aspects of good management?
Whereas central bankers rely on a vast array of data to inform their
decisions on just one variable in pursuit of a single target, the fiscus
carries the entire matrix of social, developmental, economic and political
aspirations of the nation. It is no surprise that the budget process is
characterised by an information overload and frequently by a woefully
inadequate quality of data and analysis.
We are getting better at this but there is still more work to be done. Good
management is very largely about good information and organising that
information in accessible digestible ways.
To return to Ronald Coase. Let's assume that we have a small river, a woman
upstream who makes leather and a man downstream who catches fish in the river.
In making the leather the woman introduces a small amount of pollution into the
river. The level of pollution affects how much fish the man can catch. Let's
assume that they both have equal bargaining power and there are no transaction
costs. Coase's theorem argues that property rights over the river could be
ascribed to either party to achieve the same social outcome. This is
counter-intuitive.
Coase argues that if we ascribe rights over the river to the man, he would
charge a fee to the women to use the river. For every rand he lost in fish he
could increase the fee to the woman therefore taxing the pollution. Conversely,
if we ascribed property rights to the woman, she could charge a fee to the
fisherman. If she pollutes the river her income would drop.
Even in this very simple example we can list all the things we need to make
this scheme work. First, we need someone to be able to monitor the amount of
fish caught or leather produced and pollution introduced into the river. We
need a legal process to draw up a contract which regulates this transaction. We
need an arbiter to decide on matters when a dispute arises. We need the rule of
law to enforce contracts even between private citizens.
If we take modern, complex societies, globalising economies, diverse
communities, wildly differing bargaining strengths and imperfect markets, the
role of government becomes incredibly complex and difficult. In taking
decisions on even simple matters, the stakeholders, interests and factors that
have to be taken into account requires the wisdom of Solomon. And I will be the
first to admit that we do not always get this right.
The prospects for successful management can be enhanced by improving the
alignment between the accountability chain, institutional governance,
managerial incentives and public policy objectives. Alongside these aspects of
the internal organisation of public services, there are also challenges in the
interaction of public and private sector arrangements and in the interplay
between national and international trends.
These dimensions of public policy add even more complexity to the discipline
of public finance and to the practice of public policy making and
implementation. The tidy division of the world into public and private sectors
and domestic and foreign affairs has given way to much more complex
institutional arrangements and hence a more elaborate intellectual
apparatus.
To demonstrate just one of these complexities, let me give you an example
from a public-private partnership (PPP) we have to equip and maintain the Nkosi
Albert Luthuli Hospital in Durban. Here, the private sector installs the
equipment in the hospital, maintains the equipment and replaces items when they
break or malfunction. Let's assume a light bulb breaks. The hospital has to
inform the company that a light bulb has broken; the company sends someone to
do a damage assessment and then sends someone else to fix it. If the bulb is
out of order for an extended period, the hospital can impose a penalty on the
private contractor. This particular PPP has won international awards.
However, for the deal to work it requires a contract with detailed
specification on what equipment is needed and what service standards are
required. The contract must have penalties for transgressions of the agreement.
It then assumes that the hospital has a system to track what breaks down when,
and when it was reported to the private company. It then needs to track when
the problem was fixed. If there was a delay, the legal department needs to be
contacted to impose a penalty on the company. If the company disputes this, it
goes to court. The court has to decide on costs based on the integrity of these
systems of tracking information. All of this entails huge transaction costs and
a level of sophistication that often does not exist in a public hospital.
In theory this PPP has the potential to improve the standard of service in
the public sector. Managing complex transactions is not easy and fraught with
difficulty. It often takes decades for the institutional environment to drive
down such costs, to make them routine and only then is it likely to see greater
use of such partnerships.
The great advantage of traditional government arrangements or procedures is
that they are simple, lines of authority are clear, rules and procedures are
documented and familiar and the annual budget process provides a transparent
and unambiguous assignment of resources to public purposes. Resources are
raised through taxes which have the great advantage to the fiscus that they are
mandatory. But co-operative or contractual arrangements with the private sector
and international collaboration in pursuit of common purposes are fraught with
negotiation difficulties, problems of trust, possible conflicts of interest,
risk, uncertainty, asymmetric power and interminable frustration.
Complexity is with us and so we have to find a way through the barriers of
misunderstanding. Unless work is done on the terms and conditions of agreements
and the appropriate financing or pricing arrangements, co-operation will not
happen.
Consider, for example, the challenge of mobilising private finance in
support of public investment goals, small enterprise development, low-income
housing, economic infrastructure, social and community investment, student
loans, small farmer support programmes.
The Financial Sector Charter sets aside large amounts of money for these
purposes but the institutional arrangements are taking considerable time to be
developed. The delays are partly about complexity, partly about trust, partly
about misunderstandings. Well-structured PPP do take some time to design and
negotiate. It would help if there were greater understanding of the basic
agency and incentive problems. One simple idea that perhaps deserves greater
currency is the idea of competition between service providers for a share of an
agreed public resource envelope, based on defined measures of performance. A
loan guarantee fund or co-financing arrangement for example, can be rationed
between participating banks on the basis of agreed performance criteria thus
avoiding the overwhelming transfer of risk of failed schemes to the fiscus that
characterises so many policy lending initiatives that have a redistributive or
poverty-focused objective.
In managing these and other kinds of partnerships and more generally in
modernising the public service delivery, it is necessary to guard against
unnecessary complexity.
The modern world offers many opportunities for public finance innovation and
for new kinds of partnership with the private sector and across national
boundaries. There are enormous benefits from getting these reforms right. But
getting them right means keeping them simple so that as reforms proceed we have
more control, know more about what is going on, have a better understanding of
how information is used and contributes to the growth of the market
economy.
Again, thank you for this award. I'm still no clearer on whether I deserve
it.
Issued by: National Treasury
5 October 2006
