Corruption Matters - June 2015 - Issue 45

To sign or not to sign; that is the question

by Dr Robert Waldersee, ICAC Executive Director, Corruption Prevention

Line of visibility

The ubiquitous signature is at the heart of bureaucratic control. Often derided as red tape, signatures formally confirm that a decision has been carried out in accordance with requirements. Additional signatures indicate managers have checked the approvals and the recommendations of those below. Surely then, more signatures means more control, right?

Arthur Laffer recently visited Australia. Laffer is most famous for a curve in economics that bears his name – the Laffer curve. At its most simple, the curve argues that no tax could be raised with the tax rate set at either 0% or 100%. So, at some point, further raising the tax rate paradoxically decreases the revenue collected. The optimum revenue raising tax rate must be greater than 0% and less than 100%.

Could a parallel be drawn with signatures? Clearly, with zero signatures, chaos would reign across government. Public officials could use taxpayer money for whatever they wanted. Accountability would be non-existent and government could not function. But the other extreme is no better. If every public official only checked the signatures of other public officials, government would cease to function. Accountability is destroyed as no one has final ownership of any decision. With signatures, as with tax rates, at some point more produces less.

Why the paradox?

How can more checks reduce assurance? There appears to be three main reasons why adding more signatures might produce a decline in control after a certain point.

First, there is the work load created by increasing the number of signatures on any given item. When additional signatures are required, or restrictive delegations are adopted, the effect is to increase the workload of more senior staff. Someone has to check and approve for every signature.

In one local council examined by the Commission, delegations were reduced to $2,000, effectively pushing most decisions up to the level of general manager for final sign-off. The sheer volume of signatures required of a small number of managers means that the managers cannot give the time necessary to each item; the checking and assurance represented by the signature becomes somewhat illusory.

At some point, managers who are overloaded by signature demands will just tick an approval and flick over to the next item in line. One organisation was able to measure the time taken for the electronic procurement approvals by its managers – it was between 10 and 15 seconds per item. Yet managers who tick and flick approvals have been central to allowing the corrupt conduct exposed in numerous ICAC inquiries to occur. They miss key bits of information that would have revealed that all was not right.

A second effect of increasing the number of signatures is to dilute accountability and, along with it, the diligence of the signatories.

In one process brought to the attention of the ICAC, an agency requires over 50 signatures. Which of the 50+ signatories can be held to account for the correctness of the final decision? Most will sign knowing that there will be more signatures to come; that is, they cannot really be held accountable. While they may still do their job well and may check to some extent before they sign, in the end the agency could have traded a lot of accountability for very little assurance by adding so many signatures to the process. 

Finally, adding more signatures can have the effect of separating the decision to approve from the information needed to make the decision. One manager told us that he signs-off on car usage of his staff who are spread across the state. He has no real idea of what he is approving and no way of finding out.

Either sufficient information must be moved to the decision point or the decision point moved to the information. Yet the opposite often happens. As the final signatures are located further up in the organisation, the signatory becomes less likely to know the details of what they are approving.

As the number of signatures passes an optimal point, tick and flick behaviour is encouraged, accountabilities are eroded and the signatory is less informed about what they are signing – not to mention the system becomes clunky and slow, which creates its own costs and perverse incentives.

So what’s the alternative?

If more signatures is not necessarily the solution, then what is?

One is to strengthen the accountabilities of frontline managers – the first signatory. In one financial services organisation, a random selection of files from frontline managers is randomly allocated to managers at the next level for review. Not only is the decision reviewed but so is the thoroughness of the decision process. The process is then replicated, with a random selection of the reviews themselves reviewed by more senior managers. Rather than additional signoff, there is effectively a forensic examination of the first signatory.

A similar effect is created by a cluster secretary, who reportedly will go straight to the person who signed first in some matters, and talk to them directly. Accountability is, therefore, strengthened and information flow improved.

Rather than adding signatures, managers may improve control by better designing work processes and information systems. Tight processes can constrain inappropriate exercise of discretion in a multitude of settings – from recruitment to inventory management to supplier engagement. Ownership of risky processes can be clearly located in manager roles. Data can be gathered and structured in such a way that unusual activity can be detected – from excess car usage, flow on engagements of contractors to contractor variations per contractor.

The control environment is broader than signatures. To improve control by adding more signatures is simplistic at best, and may eventually become counterproductive. In many situations there are alternatives to the addition of signatures to control discretion.

To sign or not to sign

back to menu page