Analytics on the workforce and wagebill • govhr

library(govhr)
library(dplyr)

## 
## Attaching package: 'dplyr'

## The following objects are masked from 'package:stats':
## 
##     filter, lag

## The following objects are masked from 'package:base':
## 
##     intersect, setdiff, setequal, union

Analytics

Once data are harmonized, analytics helps governments answer specific policy questions: How compressed is the wage structure? Is the workforce growing or shrinking? How is headcount distributed across groups?

govhr provides functionalities that enable users to generate analytics, along with companion plotting functions to visualize them. Broadly speaking, we differentiate between analytics on the workforce and the wagebill. Workforce analytics focus on the composition, growth, and distribution of personnel, while wagebill analytics examine patterns and trends in compensation (e.g., base salaries, allowances).

The following examples illustrate how to leverage govhr to generate analytics.

Workforce analytics

When analyzing the workforce, a first order concern is understanding time trends in headcount. One way to do this is to compute the total headcount by reference date, allowing the user to visualize the trajectory of the public sector workforce over time.

The function compute_time_trend and its companion plotting function plot_trend allow users to quickly generate this type of analysis.

# total headcount by reference date
headcount_trend <- compute_time_trend(
  govhr::bra_hrmis_personnel,
  group_col = "ref_date"
)

plot_trend(headcount_trend, group_col = "ref_date")

You might also be curious about how time trends for headcount varies across groups, such as gender. Modifying the group_col argument makes that possible. Note that implicitly the function will still take the reference date into account.

# total headcount by reference date and gender
headcount_trend_gender <- compute_time_trend(
  govhr::bra_hrmis_personnel,
  group_col = c("gender")
)

plot_trend(headcount_trend_gender, group_col = c("gender"))

Movement

Beyond the stock of personnel, governments are often interested in the flows of personnel into and out of the public sector. The compute_workforce_movement function allows users to compute the number of new hires and separations by reference date, while the companion plotting function plot_movement visualizes these flows.

movement_hire_count <- compute_workforce_movement(
  govhr::bra_hrmis_personnel,
  movement_type = "hire",
  measurement_type = "count",
  group_cols = "ref_date"
)

plot_movement(
    movement_hire_count,
    movement_type = "hire",
    measurement_type = "count",
    group_cols = "ref_date"
)

Again, if you are interested in how these flows vary across groups, you can modify the group_cols argument to include additional grouping variables.

movement_hire_gender_count <- compute_workforce_movement(
  govhr::bra_hrmis_personnel,
  movement_type = "hire",
  measurement_type = "count",
  group_cols = c("ref_date", "gender")
)

plot_movement(
    movement_hire_gender_count,
    movement_type = "hire",
    measurement_type = "count",
    group_cols = "gender"
)

Wagebill analytics

Most of the analysis presented in the workforce analytics section can be replicated for the wagebill. For example, the compute_time_trend function can be used to compute the total wagebill by reference date, while the plot_trend function can visualize that trend.

# total wagebill by reference date
wagebill_trend <- compute_time_trend(
  govhr::bra_hrmis_contract,
  group_col = "ref_date",
  measure_col = "gross_salary_lcu"
)

plot_trend(wagebill_trend, group_col = "ref_date")

A similar differentiation by group is possible, using the same grammar for the group_col argument.

# total wagebill by reference date
wagebill_paygrade_trend <- govhr::bra_hrmis_contract |>
    filter(!is.na(paygrade)) |>
    compute_time_trend(
        group_col = "paygrade",
        measure_col = "gross_salary_lcu"
    )

plot_trend(wagebill_paygrade_trend, group_col = "paygrade")

Equity

Generally, equity analyses focus on whether different groups of personnel are compensated fairly. It also includes analyses of whether the wage structure is compressed, meaning that the gap between the highest and lowest earners within a group is small.

A compressed wage structure can indicate that the government is paying its employees more equitably, but it can also suggest that there is little differentiation in pay based on experience or performance. How to interpret the findings of this equity analysis depends on the policy preferences of governments and their public sector workers.

# wage compression: ratio between the 90th and 10th percentile of gross salary,
# tracked over time
compression_ratio <- compute_compression_ratio(
  bra_hrmis_contract,
  measure_col = "gross_salary_lcu"
)

plot_compression_ratio(compression_ratio)