WGEA Vs GREAT: Actionable Gender Pay Insights

The Workplace Gender Equality Agency (WGEA), and its findings, triggered headlines across the nation in February by publishing data on gender pay gaps of Australian organisations with more than 100 employees. The headlines were damning, and the naming and shaming triggered a range of responses from various stakeholders ranging from recognition of the need to do better, to public record attempts to explain away the findings with criticism of the WGEA methodology (which in some cases did not come across well). It is no secret that the way that the information is collected, analysed and reported reveals a real problem, nor that it does so in an oversimplified way that neither tells the full story, nor helps organisations identify specific areas for further consideration and action. While this appears to be an intentional method of inspiring action, this insight focuses on ways that organisations can get genuine, specific insights and develop action plans to address gender equality in remuneration. GRG believes strongly in this outcome and can offer practical support across all role levels and organisation scales.

Why the Headline Number is Both Right and Wrong

The headline gender equality number in some ways cannot be wrong, in that it places all employees on the same footing and looks at what people take home in remuneration, by gender. For most organisations, it tells the truth that males generally take home more remuneration than females (assuming binary gendered analysis). The implication of this is that remuneration is “unfair” and “unequal”. While this is certainly true in some places and cases, the explanation for most organisations is much simpler, based on GRG’s experience offering advanced analytics, and considering the WGEA data. There are a number of factors that may be driving these outcomes:

1. Senior role samples tend to be male dominated,
2. Female dominated job functions/job families tend to be lower paying (such as early education, nursing, PA/EA or administration), while higher paying job functions/job families (such as engineering or IT) tend to be male dominated, even when the roles are graded to the same level,
3. Part-time incumbent populations are more likely to be female dominated, which also has consequences for performance assessment, career progression, and bonus calculations.

As a result, it is true for many organisations to say that they pay the same remuneration to the same roles regardless of gender, and that the headline numbers are skewed by a small number of executives, and demographic distribution among roles or functions that pay differently. However, there are three important considerations that need to be addressed when such assertions are made, or such a belief is used to explain away the headline numbers:

1. How does the Board/management know that there are no other explanations or areas of concern unless it has looked in detail, using a wide range of methods to draw out potential areas of bias or improvement? This is the focus of the remainder of this Insight.
2. The headline outcome will not show parity until the broader social challenge of correcting under-payment of traditionally female dominated job families and functions is addressed – market benchmarking will perpetuate this problem indefinitely unless affirmative action is taken. This is a complex problem that will not be explored further in this insight but is within the power of every employer to start addressing.
3. The headline outcome will not show parity until gender diversity in senior, executive, and Board roles is addressed. This challenge has a range of potential solutions that will not be explored further in this insight as it is already generally well documented.

Most Companies Have Bias-Free Frameworks to Ensure Equality … in Theory

It is also true that most organisations that are the subject of WGEA reporting are sufficiently large that they have formal remuneration frameworks, and access independent external benchmarking data, which in theory assures that roles are paid comparably, regardless of gender. Unfortunately, this is not sufficient to ensure true remuneration gender equality, and both conscious and unconscious bias can still create opportunities for action. Taking a deep dive into advanced analytics can show whether there are problems, and just as importantly, highlight areas where there is a risk of perceived bias, that should be monitored, investigated, or explained to ensure that a genuine problem is not being observed. For example, the outcome of affirmative action to develop and promote female incumbents into higher paying male dominated roles is often a temporary observed bias towards lower positioning of females against a role benchmark. This is because most “neutral” remuneration frameworks involve a “pay range” around a benchmark, through which incumbents progress as they develop from “new and inexperienced”, through to “experienced high-performer”, and ultimately promotion up into a new pay range or band. While this can be observed in a properly operating neutral framework, it is not uncommon to find that in fact experienced female incumbents do typically receive slightly lower remuneration against neutral benchmark market data for the role, which can often be an indication of genuine, if subtle, bias.

Taking a Closer Look and Getting Assurance

The majority of organisations appear to have cause to take a closer look at remuneration equality, based on the WGEA findings. The only way that a Board or management team can be certain that it does not have a genuine problem, or to properly target plans for action if it recognises it can address potential issues, is to take a deep-dive into a range of remuneration related data and practices. The following outlines a few of the key methodologies that can be used to better diagnose areas of potential risk, and target action:

1. Take a closer look at performance ratings as these can impact a range of remuneration outcomes, such as:
   a)  “target position in range”,
   b)  annual increases in pay reviews, and variable remuneration or bonus outcomes,
   c)  opportunities for development and promotion that lead to higher pay.
   Observing that females more often receive lower ratings than males is likely indicative of a problem, and this can be done at group, business unit, or even location levels.
2. Use “compa-ratios” (the percentage an incumbent receives of a neutral benchmark for a role using independent external market data) to explore whether you really do pay the same money for the same job, and if not, where some further investigation may be warranted.
3. Leverage grading systems and job evaluation systems to examine whether market benchmark data is contributing to the headline outcome, and to disentangle outcomes for roles with the same title but operating at different levels.
4. Except in the case of award and roles covered by an enterprise agreement, most organisations apply a “pay range” to jobs. Therefore, while it is fair to say that in a broad sense, most organisations pay the same amount (within a range) for the same role, usually there are in fact variations. Closely inspecting the distribution of remuneration for the “same job” can often reveal surprising results in some, but not all, parts of the business, which is an opportunity for targeted action.
5. Most remuneration frameworks include job classification systems to assist in benchmarking, which allow roles to be grouped by title or job-match code and checked against the benchmark market data applicable to them. Some organisations appear to show a trend in the “compa-ratio” of remuneration for females relative to males, so while the same benchmark and range applies, how individuals are positioned relative to the market benchmark for the role can sometimes reveal subtle bias.
6. Year on year movements (annual increases) are sometimes a surprising source of bias. These increases tend to be more discretionary than benchmarking, and allow opportunity for bias to creep in. Checking that annual movements do not show a gendered bias, or perhaps a positive bias, is likely to be critical to reducing the headline gap.
7. Variable remuneration such as short term incentives or bonuses are also often the subject of discretion and preferences devolved to levels of the organisation that may not be as objective as benchmarking processes. Examining the distribution of variable remuneration can reveal surprising biases in some organisations, although some of this can be driven by benchmark data indicating that female dominated job families tend to attract lesser variable remuneration. Even in this case, this is something that organisations can address.

Gender pay equality is a classic example of something that should be the subject of independent assurance, to ensure that the parties that may either be responsible for applying bias, or that may be concerned about the consequences of oversight of such bias, cannot influence the findings of a deep-dive into the issue. While self-audit has some value, independent audit is likely to be viewed by stakeholders as superior.

Examples

The following outlines some examples of outputs that can be used to analyse practices and identify areas of further investigation or action: