The difficult trade-off between clean audits and innovative government

A regular news items at this time of the year is the Auditor-General’s announcement of provincial audit outcomes.

The holy grail is a “clean audit”. This year the Western Cape achieved 12 clean audits out of 13 departments, an unprecedented achievement in the history of South Africa’s democracy.

But what does this actually mean?

One of the first questions I was asked following the announcement of this year’s audit results was what action I would take against the Minister and Head of the only department that did not get a “clean audit”. The answer is none. Because it is wrong to assume that this implies financial mismanagement or corruption. It does not.

It took me years in government to learn that South Africa is the only country in the world that has developed the concept of a “clean audit”. And there is a good reason why other countries have not. Trying to achieve a “clean audit” can actually become a stumbling block to service delivery. It also puts a brake on innovation in government.

At face value, this statement sounds outrageous, so let me unpack it.

In the rest of the world, an audit outcome measures the state of an organisation’s finances and falls into one of three categories: unqualified (no problems), qualified (some problems) and “disclaimer” (serious problems). This form of oversight is essential to good governance.

An unqualified financial audit signals that the department’s finances are in order. I am pleased to say that, once again, the audit outcomes of Western Cape government departments’ were 100% unqualified.

However, in South Africa, a government audit investigates much more than a department’s finances. It also includes two other categories:

• Whether a department achieved its “predetermined objectives” according to set criteria; and

• Whether it complied with all laws, regulations and directives.

A clean audit requires attaining the set standard in all three categories. Again, at face value, that sounds like a laudable goal. But there are unintended consequences that sometimes nullify the intended gains.

For example, the requirement that a department achieves its “pre-determined objectives” means that we avoid “stretch targets”. We set the bar a little lower than we should.  In my experience, we would achieve much more by narrowly missing an ambitious target than easily achieving a lower one. But this approach does not get you a “clean audit”.

In addition, pre-determined objectives (PDOs) must pass the SMART test. They must be Specific, Measurable, Achievable, Realistic and Time-bound (a tall order given hundreds of PDOs).

When it comes to “compliance with all laws, regulations and directives”, we land in a morass of red tape. If we had been held to this standard in 2009/2010, we would never have delivered the World Cup on time. It was only because so many “deviations” were authorized, that we were able to get through the interminable procurement procedures required. And even if we had followed every single procedure and prescript to the last detail, we would not have been able to prevent the collusion among construction companies that allegedly occurred.

The “compliance” criterion sometimes lands us in classic Catch 22 situations, where we are damned whether we “do or don’t”. There is, for example, a national directive that senior posts must be filled within 90 working days of becoming vacant.

To achieve this we need to advertise the post, sift through hundreds of applications, draw up a “long list”, get a panel to prepare the short list for interview, conduct a full day psychometric tests on the shortlisted candidates (another national requirement), prepare and conduct the interviews.

However, another national directive requires every senior appointment to be “vetted” before an appointment letter can be issued so that we can detect fraudulent qualifications, criminal records, etc. It is difficult – and often impossible – to fulfil all these requirements within the 90-day period.

We depend on educational institutions and other agencies to verify qualifications, criminal records, etc. and this process often takes a long time.

So we have a stark choice. Either we fill the post in time, or we comply with all the process requirements and the vetting procedures  --   which can only begin after the interviews, when a candidate is recommended. A few years ago, we almost forfeited a number of clean audits because of the double bind of the “90-day” rule.

We face similar anomalies all the time. There is a requirement, for example, that any person contracted for any purpose must be registered on a supplier database. A few years ago, when we were in the process of updating and changing the format of the supplier database, we had to procure the services of a caterer for a cheese-and-wine function.

The only person that complied with the registration requirements was a caterer in Beaufort West, which meant that we would have to incur the additional cost of an air flight in order to comply with the audit criteria. I lodged a complaint with our own “red tape” unit, and the issue was resolved.

This year we came close to losing our clean audit in the Department of Cultural Affairs and Sport (DCAS) because of 7 missing library books. This example is instructive, and illustrative of many others.

There are 6,6-million books in the libraries managed by DCAS. A National Treasury finance instruction deems these books to be assets of the state, so that when the AG audits libraries, he takes a random sample of 210 books (out of 6,6-million). If seven out of the 210 are missing, the department loses its clean audit, as there will be an overstatement of the assets which is considered to be material to the audit outcome.

On the scale of our library operation, books sometimes go missing, either because borrowers have not returned them on time, or because they have been incorrectly filed.

This year we almost failed the “missing book test” leading to a frantic search for the seven missing books (of which five, fortunately, were located). With only two missing books out of the audit sample, DCAS was in the clear on this score. There are dozens of similar “tick-boxes” that require compliance. It is an exhaustive (and exhausting) process.

Using these examples, it is easy to see why the single-minded pursuit of a clean audit can curb service delivery. If the Department wanted to ensure that there was no chance of failing the “missing book” test, they would have to close down libraries for a month, conduct a full stock-take, confirm the asset register, and keep the libraries closed until the end of the audit period.

Obviously this is not feasible. If we had to choose between keeping the service running or forfeiting a clean audit for seven missing books, we would choose the former.

The learner transport system offered by the Western Cape Education Department provides another example of what getting a “clean audit” involves. The transport service is provided mainly for learners who live in remote parts of the province, five kilometers or more from the nearest public school, without access to other forms of transport.

There are various pick-up points along the farm road networks, to benefit as many qualifying children as possible.  Because roads in remote areas are affected by many different factors, new routes must sometimes be devised.

During the audit process, however, the AG’s main concern is whether the route is the shortest one possible. If it is not, for whatever reason, it is deemed to constitute “irregular expenditure”, as was the case two years ago. However, if we slavishly prioritize the shortest route, some learners would not have transport to school at times (if roads are flooded, for example).

And then there is the cost:  over and above the approximately R100-million annual fee the AG charges the Western Cape for the service, a large number of competent professionals spend a lot of time to ensure that we get clean audits. Some spheres of government, not as well endowed with the relevant capacity, spend additional millions on consultants to prepare for the audit.  Sometimes the cost:benefit ratio is out of kilter. As that wonderful Afrikaans saying goes:  the cabbage is not worth the sauce.

Governments have the complex task of balancing various priorities.  We have to spend public funds wisely and well, in order to create an enabling environment for economic growth and job creation, while providing the essential services people cannot provide for themselves such as clean water, sanitation, education, health care, among others. All of this must take place within very tight budgets. We often have to make difficult choices between competing priorities.

And this does not always fit the “tick box” system. Officials just stick to the formula, even if they know there is a better way of doing things. And they avoid any risks, in case it leads to a qualified audit.

We acknowledge the crucial role of the AG in ensuring good governance, and we regard the AG's office as our partner in achieving this goal in the Western Cape. However, we  also believe it is time for our auditing system to become more nuanced. We need to introduce “value for money” auditing, which determines whether the choices we make provide the best value for public money in the given circumstances.

Of course, this form of auditing is much more complex, and requires subjective judgments. These are bound to be controversial at times. But it is the only way to address the absurdity of the quest for “clean audits” undermining the overriding goal of effective (and cost-effective) service delivery.

This article by Helen Zille first appeared in Inside Government, the online newsletter of the Premier of the Western Cape.