The paradox of performance pay

Performance pay has no place in the public service, whether for agency heads or APS-level staff

Leaders often define themselves by the issues on which they take a stand. My 7½ years as a secretary in three departments of state spanned an era when performance-based pay became the order of the day; a fashion I resisted. The essence of my arguments is that performance pay:

■ Is at odds with public service culture.

<i>Illustration: Matt Adams</i>
Illustration: Matt Adams 

■ Ignores the complexity of how the public service actually works.

■ Is bad for morale and teamwork.

■ Gives senior leaders an excuse to avoid real leadership.

During my public service, I came to appreciate the importance of rituals, symbols and words. So it may come as no surprise that I believe the term ''human resources'' indicates a mindset that treats people as human capital, assets, units of production and so on, with inherent control connotations. The term should be outlawed in favour of the more positive notions of people and performance.

In my view, performance-based pay experiments in the Australian Public Service have all been abject failures in the eyes of the people affected by them. Given the complexity of much of the public service, success or failure are mostly shared outcomes. That's because responsibility for meaningful bundles of work can rarely be made coincident with individual responsibility. This factor alone bedevils performance pay. Performance appraisal and pay assumptions usually include that:

■ Evaluation covers performance over a (normally) 12-month cycle, not just the period of recent memory.

■ Evaluators are consistent with one another or equilibrated through a moderation process.

■ Evaluators apply a consistent ''objective'' standard between employees.

■ Individual employee contributions can be distinguished from the contributions of other managers and workers.

These beliefs usually arise from the command and control school of management.

Douglas McGregor developed Theory X and Theory Y in the 1960s to describe what he discerned to be two very different approaches to workforce motivation. Importantly, Theory X and Y are not different ends of a continuum as commonly thought.

Based on traditional views of direction and control, Theory X asserts that:

■ The average person has an inherent dislike of work and will avoid it if they can.

■ Most people must be coerced to contribute toward organisational objectives.

■ Most people prefer to be directed, wish to avoid responsibility, have relatively little ambition, and want security above all.

Theory X managers see their job as being to ''motivate'' their employees, believing carrots and sticks are the only effective way to get things done. Frederick Herzberg proved 30 years ago that motivation is an internal construct and you can't actually motivate others to do much of anything other than what they want to do, or what they perceive to be in their self-interest.

This might be illustrated by how supervisors respond when an employee approaches them about a stuff-up. The carrot-and-stick school tend to shoot the messenger, assign blame and punish. So employees learn not to confess further errors, potentially leading to disastrous consequences. Those with a more positive view thank their staff for drawing attention to the problem and work with them to fix it. They get told of future problems as they emerge and have the opportunity to resolve them before things get too far out of hand.

I believe the vast majority of APS staff would not see performance pay as being in their self-interest. The rewards (even for secretaries) are relatively small, the ranking system rankles because many good performers not given the top rating think they have been short changed, and the system is not regarded as fair. Leadership and management policies and practices provide an insight into how leaders regard their employees; The book Punished by Rewards (1993) argues the case against the carrot-and-stick approach.

Performance pay acolytes (usually in an unstated or unrecognised manner) consider that a reservoir of withheld effort must be coaxed or coerced out of people. This is the underlying premise for incentive pay schemes and/or a manager's efforts to motivate and control their staff.

The reality is that there are not many bad people in the workplace. What's interesting is that so many executives shy away from dealing directly with poor performance and unacceptable conduct in the workplace, especially given the way those affected by these inappropriate behaviours feel about it. One might therefore expect this aspect to feature in performance-pay regimes, but even there it seems to be the exception rather than the rule.

Most APS performance-based-pay programs follow a three-step process:

■ Determining the approach: what's to be done, by when and to what standard.

■ The performance period: normally the calendar or financial year.

■ The performance review: the boss undertaking an ''objective'' evaluation of the subordinate's performance.

Typically, evaluation involves a boss/subordinate discussion followed by the assignment of a numerical rating, sometimes with forced rankings according to the classic bell-shaped normal distribution curve. (By contrast, the Defence Department's system was based on incremental advances up the pay scale, without forced rankings). The outcome usually has a remuneration implication covering pay at risk, merit pay, bonuses and judgments about contenders for promotion. Responses to ''why do it?'' usually say:

■ As a basis for differential pay/reward for outstanding performers.

■ To provide performance feedback to individual employees.

■ To identify candidates for promotion.

■ To foster communication between supervisors and subordinates.

■ To motivate employees.

As I recall it, performance pay was introduced as a back-door way of lifting remuneration for senior staff. That could not be achieved through the Industrial Relations Commission because of the very restrictive rules applying to pay increases at that time, which the government had advocated to the commission and felt bound to adhere to. It was argued that, as employees contribute at different levels of effectiveness and effort, this could be recognised by introducing a performance-pay regime. The fact that that rationale no longer applies seems lost with the effluxion of time, though some departments and agencies have now done away with performance-based pay.

Nevertheless, the practice remains widespread in the APS because so many managers believe in it. They believe it works without consciously and critically analysing the assumptions behind the practice. And they believe in it, despite the fact that 90 per cent of managers consider the approach to be unsuccessful.

Some commentators argue that the practice continues despite evidence to the contrary because it relates to the manager's need to maintain control, or the illusion of control. Peter Scholtes, who has researched and written extensively about performance, appraisal and pay, argues that such a performance ''management'' regime is inherently the wrong thing to do because three faults are common to all variations on the theme:

■ It doesn't work.

■ It's wrong to focus only on individuals or groups, because most opportunities for improvement involve systems, processes and technology.

■ Performance ''management'' is judgment, not feedback; it's a hierarchical dynamic.

Even when well intended, ratings are judgmental and related to control of the person being evaluated based on the presumption that the person's inadequate contribution is separable from any systemic origin of poor performance. Where such schemes are in operation, most individuals or groups will work towards optimising their performance, regardless of its effect on the system. Creative accounting, goal displacement, withholding information, reduced quality at the expense of more output, individual visibility which discourages cooperation and other gaming strategies are the perverse results of such a perverse system. The inherent contradiction of proselytising individual performance-pay assessments while simultaneously exhorting teamwork escapes the zealots.

The method, criteria and philosophy of evaluation differs between evaluators. People are subjective and they are not at all objective about their subjectivity. Numerous factors affect favourable or unfavourable bias, including age, family and educational background, physical appearance and so on. None of these are job-related, but they do influence the outcome of performance evaluation. Where these discriminatory practices are pointed out, the people concerned deny they exist.

Performance pay can lead to patronage, subordinate sycophancy, playing and paying favourites, oiling the squeaky wheel and other inappropriate practices. Imagine the consequences if ministers wanted to be involved in the process and decisions below secretary level.

A former secretary once commented that he had assessed his senior executive service officers as all being in the top 5 to 10 per cent! Let's assume for the purpose of illustration that there is a truly unbiased performance-pay system in a typical Gaussian distribution. Half of your people will learn that they are below average: a statistical inevitability. Some may accept their fate; others will view this as proof that their manager is incompetent. Some will redouble their efforts to prove the judgment and system wrong - that may be noticed, and they may be lucky enough to be ranked above average next time − if so, someone above average last time will fill their below-average slot this time. All of this must do wonders for morale and superior workplace performance.

Then there's the case where people are told their rating by their direct supervisor before it disappears into the black box of moderation and comes out at a lower level; apart from the lack of transparency, the recipient's perception is one of unfairness and it's deeply demotivating. Those who invoke ''science'' in moderating people's scores to a decimal point are kidding.

The Orwellian named ''efficiency dividend'' has had particular effects on some small agencies (particularly those where most of their budget comprises staff costs), leading to very significant differences in pay for people doing jobs classified at the same level. The reality is that how well or badly people are paid in these circumstances often depends on how well their organisation has done in the budget bidding process. As well as the equity argument, this works against mobility and a unified APS.

The reform blueprint of former Department of the Prime Minister and Cabinet secretary Terry Moran analyses what's happened since the devolution of wage bargaining in 1997, and it's spot-on - as are the accompanying recommendations. There is a compelling case to abolish performance pay and return to centralised pay-fixing, perhaps under the auspices of the Public Service Commission.

Dr Allan Hawke is a former chief of staff to prime minister Paul Keating, a former federal departmental secretary and a former senior diplomat. This is an edited extract from an essay he wrote for With the benefit of hindsight, a collection of valedictory reflections from departmental secretaries, published by ePress at the Australian National University.