Subscribe to newsletter
10/03/2025
Every year we analyse the votes against ASX 300 company remuneration reports and explore the relationship with financial and market value factors. With 2024 strike data, it cannot be said that the record high number of 2023 strikes was an outlier. Instead, we may be at the beginning of a new voting trend. Factors previously linked to remuneration support (or, rather, the absence of support) remain relevant.
The sample for each year was the ASX 300 as at calendar year end, excluding foreign incorporated companies due to differences in disclosure and other regulatory requirements.
Conclusions from analysis:
From our analyses, we have drawn the following conclusions:
- The voting dissent in 2024 was the second highest recorded, following the record surge in 2023, suggesting that 2023 was not an outlier.
- Unlike prior years of high no votes, 2024 was not benign. Some would say this is sufficient to call the absence of remuneration support a trend. While this trend may be the new normal, more time is needed to determine whether it will persist.
- Consistent with the hypothesis that investor dissatisfaction is settling into a new normal, the median votes against were higher in 2024 compared to 2023.
- Sector-level analysis reveals a mixed trend. Some sectors experienced a 2024 decline after 2023 while others continued to see increases in 2024. Since the median is less prone to outliers than an average, these sector-specific variations may help explain why the 2024 median surpassed that of 2023.
- Companies that rank poorly in relative Total Shareholder Return (rTSR) face the highest levels of shareholder dissent. As rTSR increases, dissent generally declines. An inverse trend emerges among companies in the top quintile where votes against rises despite strong market performance.
- Companies with lower total assets received higher votes against. The median votes against in the bottom tertile is nearly double that of the top tertil.
- The highest average votes against for 2023 were observed in the information technology sector. This peak shifted to the healthcare sector in 2024. While most sectors saw higher shareholder support following the 2023 peak, some sectors still experienced a decrease.
Longitudinal trends
The increased levels of shareholder dissent observed in 2023 remained high in 2024. The 2024 average vote against remuneration reports was 11.23%, a slight decline from 12.14% in 2023. The median vote against in 2024 (4.4%) was greater than seen in 2023 (4.3%) whilst prior medians ranged between 2.6% and 3.4%.
Figure 1: Yearly Average and Median Votes Against Remuneration Report (%)
Thirty-seven companies in our sample received a strike in 2024 (40, if foreign incorporated entities are included). This was lower than 2023 (41 strikes). In the 10 years prior to 2023, the greatest number of strikes in a year was 25, observed in 2021.
Figure 2: Yearly Strikes Against Remuneration Report (%)
Strike numbers in 2024 remain high, so that the surge in 2023 shareholder dissent was not an outlier.
Trends by Sector – is there a perennial problem child?
Companies in the Information Technology, Industrials, Consumer Staples, Consumer Discretionary, Financials, Energy, and Utilities sectors experienced a decrease in their votes against while the other sectors saw an increase. Companies in the Information Technology sector experienced the largest decrease with a drop from 18.87% against in 2023 to 12.09% against in 2024.
One-Way ANOVA (Analysis of Variance) is a statistical test used to determine whether there are significant differences between the means of three or more independent groups. In this analysis, One-Way ANOVA is applied to assess whether the voting patterns vary across different GICS sectors. By comparing the variance within each sector to the variance between sectors, the test helps identify whether sector affiliation may relate to voting behaviour. Based on the results from the One-Way ANOVA we concluded there were no statistically significant differences in the vote against rates across the GICS sectors. This means the vote against rates were statistically similar between the sectors.
Figure 3: Average Percentage of Votes Against Remuneration Report by GICS Sector for 2023 and 2024 (%)
For 2024 the highest number of strikes occurred in the Industrials and Health Care sectors with 6 strikes each. In 2023 the highest number of strikes was in the Materials sector at 9, followed by Information Technology, Industrials, and Financials sectors with 6 strikes each.
Figure 4: Strikes by GICS Sector in 2023 and 2024
The Information Technology sector saw 11 strikes in 53 votes (21%) over the 3-year period including 2024. The Utilities sector received 2 strikes in 9 votes (22%) over the same period. The Materials sector had the lowest rate of strikes over 3 years with 15 strikes in 178 votes (8%).
Impact of Financial and Market Value Factors
Figure 5 shows the impact of TSR on median voting outcomes by analysing each company’s annual rTSR percentile rank against other ASX 300 companies from the same year. These percentile ranks are grouped into quintiles.
The figure shows that protest votes generally decreased as company TSR performance improved, particularly noticeable between the first 2 quintiles. From the 4th to the 5th quintile, the trend reverses with an increase in protest votes. The lowest rTSR group exhibits the highest median vote-against rate (10.05%), while the 4th quintile (percentiles 60-80) shows the lowest vote-against rate (3.10%).
Figure 5: Average Percentage of Votes Against Remuneration Report by Annual rTSR Quintile (relative to ASX300)
The relationship between negative remuneration outcomes and total assets was explored. The boxplot in figure 6 showcases the distribution of votes by total assets tertile with a box for the 1st and 3rd quartile and lines representing the minimum and maximum. Statistical outliers are represented as dots. It was found that companies in the top tertile had a lower distribution of votes against alongside a tighter spread of vote outcomes (barring outliers). Declining outcomes spread by tertile imply that companies with higher total assets faced less shareholder dissent in 2024.
Figure 6: Votes Against Remuneration Report by Total Assets Tertile in 2024
The same procedure was used to measure the effect of return on equity (ROE) on shareholder dissent in 2024. Consistent with previous findings, no definitive relationship emerges between ROE and votes against remuneration.
Methodology notes
We analysed the influence of various financial factors on say-on-pay resolutions at AGMs held by ASX 300 firms in 2024.
The sample of ASX 300 constituents were taken as at end of each year on a rolling basis. This method differs from previous years where current year ASX 300 constituents were chosen as the base cohort with voting outcomes relating to that cohort’s performance in prior years. This change in method allows for analysis of trends in the ASX 300 over the years using voting data relevant to companies included in the index for that particular year. The previous method analyses historic trends in the current year ASX 300 cohort.
Companies incorporated outside of Australia were omitted due to inconsistent reporting standards. Companies that do not disclose/hold voting on remuneration were omitted.
Voting data relates to the calendar year in which results were published with corresponding financial data referring to the relevant financial year. This may introduce uneven exposure to varying market trends due to differing reporting dates but ensures consistency between voting matter and voting outcomes.
Company information and financial metrics were sourced from LSEG Workspace, while voting outcomes were sourced from Diligent Market Intelligence and manual collection.
We used an ANOVA test to compare voting outcomes across different GICS sectors. For all hypothesis tests, a significance level of 5% was employed. Our sample size was 243 after removing records with no financial data.
We selected our regression model using a “best subsets” approach to identify the optimal combination of independent variables while excluding interaction effects for ease of interpretation. The response variable was the percentage of votes against say-on-pay proposals. Unlike previous years, we did not apply a logit transformation, as a comparison showed that logistic regression did not improve the model’s overall interpretability or R-squared.
After performing best subsets model selection, we arrived at the following optimal linear regression model using the adjusted criterion for goodness-of-fit:
VotesAgainst = β0 + β1*TSR + β2*Energy + β3*Healthcare + β4*log(TA)
The best subsets selection procedure involved identifying the optimal model for each number of predictors, ranging from one predictor up to a full model containing all 13 possible predictors (not including interaction terms).
Although this model was selected as optimal using the ‘best subsets’ approach, the overall regression model was not statistically significant, with a p-value of 0.4705—well above the 0.05 threshold. However, upon examining individual variables, TSR had a p-value of 0.0664, and log (TA) had a p-value of 0.0869. While both are slightly above 0.05, they are still below 0.1, suggesting that these two financial measures may be related to the vote-against rate. Additionally, none of the sector indicator variables were statistically significant. Based on these results, we cannot conclusively state that the vote-against rate is significantly related to financial measures. This finding is somewhat different from previous years, in terms of both the overall model significance and the p-values of individual variables, but the general trend remains consistent: financial measures appear to be linked to the vote-against rate. This discrepancy may be due to our analysis including only 2024 data. Prior studies incorporated data over a decade, however this may induce a selection bias when creating a sample of companies who ended in the ASX 300 after 10 years.
Back to all articles
