Submission to the APS Hierarchy and Classification Review

By Tobias Beckmann

Executive Summary

The Australian Public Service (APS) must be a nimble and dynamic organisation in order to serve Australians, Parliament and governments for years to come. However, in many ways the APS is constrained by a classification system that, while perhaps useful to manage a bureaucracy in the 1980s, is no longer fit for purpose. In particular, I think that the APS is held back by a classification system that has too many classification levels, leads to an inability to think outside hierarchical lines, and makes it more difficult to attract and retain talent.

To address these issues, I propose that the APS Hierarchy and Classification Review recommend a significant consolidation of classifications to a five-level classification system. This would be combined with an uncoupling of the classification system from the system of hierarchical management of staff. Together, these would help attract and retain staff, and ensure that the APS has the diversity of skills and the flexibility to meet the challenges facing Australia.

I would be happy to discuss any of these matters further with the Review.

Introduction

The APS has a quite bureaucratic structure, where individuals’ position in the hierarchy, as well as their responsibilities and pay were rigidly defined by their classification level. But why have a classification system at all?

From my perspective, a classification system is useful for administrative control of and for enforcing clear lines of decision-making and financial delegation. It provides a simple identifier when transferring or seeking promotion across agencies, as ‘merit’ has been established in terms of being able to perform at a particular level. Furthermore, it provides transparency and accountability by publicly identifying the responsibilities and wages that an individual APS staff member earns.

However, in my experience the classification system is an imperfect proxy of strengths and expertise, which leads to the following problems:

The classification system is overly complex, creating artificial barriers for staff progression and movement.

The classification system unnecessarily combines hierarchy along with pay and responsibilities, meaning that APS staff are too often viewed simply in terms of their ‘number’, with too much weight placed on seniority.

The classification system is unnecessarily rigid, making it difficult to attract and retain talent and discouraging diversity within teams.

What are the problems with the classification system?

An overly complex system creates artificial barriers

The APS currently has 13 classification levels, which I understand is high relative to other private organisations, and indeed other public service organisations both in Australia and abroad. There is little need to have so many classifications, particularly if the managerial line is considered separately from the classification system.

A complex system creates barriers to progression

There are a significant number of roles within the APS where the same activity is performed by a variety of different classification levels. For example, an APS4 and an EL1 can often be doing a very similar task, obviously with difference in the quality of their output, their need for oversight or their speed. However, between each individual classification level, these differences are points on a spectrum and any strict delineation between neighbouring classifications is generally quite slight.

But what this means is that an individual who begins as an APS4 will then have to apply for a promotion at each point of progression, and justify how they are performing at a higher level. This significantly increases the administrative burden on the management of staff, as competitive rounds need to be advertised, interviewed for and filled to allow an individual to progress. In addition, if there is a hiring freeze mandated or there are no more senior roles available, then the staff member will be stuck at the maximum pay for their current band, even as they build their performance in their role. For example, I previously worked with a substantive APS4 who was acting (and performing highly) as an APS6, but they were unable to obtain a promotion within their current agency because there were no substantive APS5 or APS6 positions available. As a result, they left a role that they enjoyed and were passionate about, to work in a different agency with much lower job satisfaction, just in order to have security in their position.

Some agencies have (appropriately) designed their enterprise agreements so that staff are able to be broadbanded from one classification to the next, at managerial discretion. However, there are many agencies where this is not available, or only for very specific transitions (for example, when leaving a graduate program). Even in agencies where broadbanding is available, there can be hesitancy from managers to actually broadband staff.

A complex system creates barriers to moving between agencies

The APS’s overly complex classification system means that differences in pay between agencies restricts movement between agencies. The APS Review identified that the majority of APS staff had only ever worked in one agency. It also noted that one negative incentive for movement between agencies is large differences in pay at any given classification level.

For example, based on the current enterprise agreement at the Australian Institute of Aboriginal and Torres Strait Islander Studies (AIATSIS), an EL1 can earn between $95,417 and $104,666, and an EL2 between $113,391 and $126,708. At the National Indigenous Australians Agency (NIAA), an EL1 earns between $107,494 and $122,513, and an EL2 between $125,115 and $148,907. Almost all NIAA EL1 and EL2 staff are therefore currently disincentivised from transferring at level to AIATSIS, as they would be above the highest pay point at AIATSIS and not be eligible for annual wage increases, despite both agencies working on Indigenous matters.

Hierarchical thinking means that seniority dominates unnecessarily

Regardless of how it is intentioned, the APS’ classification system is often interpreted as a rigid hierarchical structure. It is therefore assumed that an EL2 will manage a project by managing an EL1, who will in turn manage some APS level staff, who might manage more junior staff again. It does not envisage having, for example, an excellent APS6 project manager leading a project, and bringing in the expertise of EL1 and EL2 technical specialists at relevant points.

As a result of this hierarchy, more senior staff are automatically given deference, as they ostensibly have more experience and responsibility, and are usually given the right to speak first (and the longest) at meetings. Some of this is of course appropriate – more senior staff will generally have more decision-making responsibilities, longer experience and greater visibility of ministers, senior executives and other areas. However, it also makes it more difficult for junior staff to question authority, to integrate diverse perspectives and to bring in appropriate technical expertise. It can also create a barrier to innovation, given that quite often seniority in the APS is actually a consequence of years served rather than ability or skill.

At its most extreme, it is rare to have someone more junior than an SESB1 brief a minister on a relevant matter. A more senior official is likely to have wider visibility of the issues at play, and perhaps greater political nous. However, they are less likely to have a deep understanding of the particular issue at hand. As such, significant hours are often required to brief the SESB1 to an appropriate level of understanding, which otherwise sits within the team but at a more junior level. As a result, more junior staff lack visibility of senior decision making and what the issues at play are, and so less able to develop their own political nous.

Rigid classifications means we don’t have the best people for the job

The current classification system is expounded through the Australian Public Service Commission (APSC) Work Level Standards (WLS) policy. However, together these policies create rigidities in staff management, making it more difficult to attract and retain talent, or encourage diversity within teams.

Rigid classifications makes it difficult to attract external talent

The rigid classification system makes it more difficult for people to join the APS from outside the public sector. The current WLS encourages staff to be able to demonstrate a wide diversity of skills and experience, some of which is only applicable to the public sector. When applying for a position, often an external hire might be less able to demonstrate this wide diversity of skills. In contrast, an applicant from within the APS is able to circumvent this issue, given that they can justify their ‘merit’ simply by quoting their classification number.

For example, I came to the APS through an external secondment, and wanted to find a permanent position. The question that each interviewer had for me was ‘but what level are you?’ If I could provide a level, then they wouldn’t need to listen to my experience or read through my application and understand the skills that I could bring to the position. This was demonstrated in that, whenever I applied for ongoing positions in other agencies, I was considered ‘not suitable’ for EL1 roles and encouraged instead to apply for APS6 positions, despite having worked for a year in the APS at this point and having demonstrated my performance as an EL1. It was only when someone who had worked with me before, and knew my skillset, that I was able to be employed at an EL1 level.

Rigid classifications encourages generalists and reduces diversity

The classification proxy also creates more homogeneity, rather than allowing for diversity of experience within teams. To achieve a promotion to the next classification level, staff are often encouraged to ‘fill in their gaps’ in the Work Level Standards. The idea is that, as a more generalist staff member, they are then more able to be deployed anywhere around the APS. This may be appropriate for some roles within the APS, but there are too many roles where a ‘policy generalist’ is employed because they are the most common tool for the job, rather than the best.

For example, when thinking about a team, a manager will often think in terms of finding “an EL1, a few APS6s and an APS4 or two”. It is assumed that, by employing an EL1 (for example), you will automatically have someone who meets all of the potential requirements that you have for the EL1 role. A manager will rarely consider the different forms of expertise that actually needs to be brought to the team: “I need someone with excellent communication skills and the ability to write briefs, two data analysts, someone with administrative experience to bring us all together”.

As a result, the team becomes a much more homogenous entity, with each individual simply points along a spectrum, being either earlier or further along their APS career journey into becoming homogenous generalists. This is in contrast with the majority of academic literature on the topic, which recognises the importance of diversity of experience and skillsets within teams.

Rigid classifications make it difficult to retain specialists in the APS

The other element to this trend towards homogeneity is that it becomes more difficult to retain specialist staff within the APS. For many specialist roles, an individual’s more narrow expertise in an area allows them to command higher wages. However, as discussed earlier, the WLS favours staff to have wide range of skills at any given level, potentially at the expense of expertise. If an individual doesn’t have that wider skillset, it is difficult to justify paying them at a higher classification level. Specialist staff are instead typically hired as contractors or consultants, which significantly increases the cost of employing the staff.

For example, suppose a specialist engineer expects to be paid at a rate typically paid to a high-paid APS EL2. However, their skillset means that they do not have managerial expertise, nor experience in corporate functions such as managing contracts and human resources. It is then much more difficult to justify employing them as an EL2, as their skillset actually aligns more closely with EL1 responsibilities. However, this means that they are not paid as much and so either the agency misses out on these skills, or employs them as a contractor.

Another related issue is that it is difficult for agencies to justify employing relatively more senior staff (e.g. SES or ELs) who do not manage other staff. This is partly because it creates an added burden on other staff through increasing their span of control. For example, an EL2 specialist is typically managed by an SESB1 and not manage more junior staff. This means that there is an increased burden on the other EL2s in the branch to manage EL1s and then onto APS level staff.

Recommendations

I will note that the APS has developed numerous mechanisms to get around these issues, meaning that none of the issues I have outlined are completely insurmountable within the way the APS currently operates. However, this shouldn’t lead to complacency. Firstly, if a classification system is so rigid that it regularly requires circuit breakers to allow for efficient outcomes, perhaps this calls instead for a more flexible system. Secondly, often staff do not know about the ways to get around the rules, meaning that these barriers do actually lead to real efficiency and financial costs.

In this light, I propose that the APS Hierarchy and Classification Review recommend a significant consolidation of classifications to a five-level classification system (or similar). In addition, the classification system should loosen the relationship with hierarchical or management control.

Consolidate current classification bands

There are a number of different ways that current classification bands could be consolidated, and the most appropriate way to do so would be based on evidence of responsibilities and delegation levels, rather than randomly grouping together similar-sounding positions. However, for an example, I offer the following five-level classification system:

Entry-level Officer (combining Trainee as well as APS 1–4)

Senior Officer (combining APS 5–6)

Executive Officer (combining EL 1–2)

Senior Executive Officer 1 (SES 1)

Senior Executive Officer 2 (combining SES 2–3)

By merging the bands, it effectively standardises broadbanding across all agencies for these levels, without requiring managerial discretion, and greatly reduces the number of points where staff need to apply for a promotion. In addition, it immediately creates greater standardisation of pay scales across the APS. To take the earlier example of AIATSIS and NIAA, merging the two ELs would create an Executive Officer band stretching between $95,417 and $126,708 at AIATSIS, and $107,494 and $148,907 at NIAA. This means that there is a lot more pay points where a NIAA staff could move to AIATSIS and not be disadvantaged by their otherwise higher pay. While it doesn’t completely standardise pay scales between agencies, it does begin to address this not insignificant problem.

Separate managerial lines from the classification system

A five-level classification system could also be integrated with a consolidation of spans of control. However, it is very unlikely that many agencies would be able to manage its entire staff with only five tiers of management. Therefore, the classification system should ideally be separated from managerial lines, so that staff can be managed by someone of an equal or a more senior rank.

This is not a new concept for the APS. There are plenty of examples of “Super EL2” or “EL2.2” positions within the APS, where an EL2 is responsible for managing other EL2s. Similar things can occur at the SES level. But rather than create an extra classification level, if the classification levels are broad enough and not tied to hierarchy then staff will not necessarily feel “demoted” when they are managed by a staff member at the same level as them.

By separating out the managerial line from the classification system, it would allow greater flexibility for agencies to maintain a specialist stream for certain professions. However, if the classifications were broader and managerial lines were separated, the relative burden is lessened. To take our earlier example, the management of the more junior staff can be distributed across a greater ‘pool’ of Executive Officer (EL1 and EL2) staff, rather than simply through the more rigid EL2 > EL1 > APS level staff hierarchy.

Outcomes

Together, these two changes to the classification system would create a more dynamic and flexible APS, with greater ability to attract and retain talent. Leaders would no longer feel that they need to only think in terms of classification levels, and instead design a management structure that suits their needs. They would have greater freedom to employ specialist non-managerial staff.

APS staff would be less artificially constrained in their career progression, with promotions to the next level actually reflecting important and significant transitions in responsibilities and delegations. The significant amount of time and resources spent on recruitment rounds would be reduced, as staff would not be compelled to apply for as many promotions to simply continue through their career.

By having wider classification bands, it is then less important to rigidly define the exact point where an external hire enters into the APS. Specialist staff are much less restrained by such narrow bands of pay that do not recognise their particular skillset. Staff that do not want to become specialists will have a much greater career path for them, as the points that they need to demonstrate generalist skillsets are fewer and better reflect the points where generalists are required.