Congressional diversity and pay: The Speaker’s data vs. Legistorm’s
Last week, Speaker Nancy Pelosi and her leadership team released the results from its 2019 Compensation and Diversity Study of congressional staff. The study, administered by the House Chief Administrative Officer (CAO), was sent out in July, 2019 to staff employed by members, committees, leadership, and House officer offices. In all, 10,356 staffers received the survey, of which 5,290 participated. The Executive Summary of the Speaker’s report can be found here, and the full report (all 189 pages, including the methodology used) can be found here.
Given that Hill staff data is notoriously messy and noisy and not one perfect dataset exists to date, I thought it prudent to compare the report’s findings on high level aspects–gender, diversity, tenure, pay, etc.–with the data I have and regularly work with. Below is what I found using staff level data provided by LegiStorm for all staff working in the House as of 3/31/2019:
Let’s start with gender. The Speaker’s report found that 54% of House staff are female (again, generalized from the 51% response rate. My data shows a 50.5%/49.5% gender split, favoring females. This isn’t too far off, and seemingly equitable in terms of gender.
But, here’s what the Speaker’s report won’t tell you. The gender split by position is far less equitable. Higher level positions trend male (especially Chiefs of Staff and Legislative Directors) while administrative-related titles are female dominated. This mirrors what I’ve found in previous congresses, and extends to what policy portfolios are given to males versus females (hint: they are not even close to the same).
Moving onto race and ethnicity, the report shows House employees are 70% white, 14.5% black, and 12.2% hispanic. Admittedly, my data isn’t as good as the first hand responses the Speaker collected with her survey —as evidenced by the 30% of staff with an ‘unclear’ race designation— but I come up with different numbers. The percentages of Black/African American and Hispanic are far lower.
Again, though, it’s important to look at these breakdowns by position. Supporting findings from previous work, in nearly all positions, and especially those in DC, minorities are vastly underrepresented. Chiefs, deputy chiefs, legislative directors, communications directors, etc., all have less than 10% of the posts filled by either black or Hispanic staffers.
Moving onto salary, the report shows an average salary of $69,379, with a bump to an average of $102,400 for staffers on committees. My data (which has every single payment given to each staffer) shows a median of about $63,000, or more than $6,000 less than the report. Here’s a breakdown of average salaries by personal office position from my data to show the wide variance.
Next, to tenure. The report lists the average tenure (years working for the House) is 5.5 years. While it’s helpful to know the average for all staff, it’s even more useful to know the numbers for each position. So, here they are.
Finally, the report gives us some of our first comprehensive looks at educational levels of congressional staffers. Little has been done on this to date, but I’ve got a big data piece coming out on exactly this topic in the coming days (or weeks if news keeps going at this clip).
I show similar results as the Speaker’s report, but I add in a little bit of nuance by adding JDs as a degree category. The big takeaways? The Hill is highly educated, especially on committees, and there are tons of lawyers.
A few last points
Importantly, both my data source and the Speaker’s survey have pros and cons, and neither should be taken as wholly representative on House staffers, particularly on race. With that caveat, I and a few others working in this space of congressional diversity, have some key questions as to the numbers and weightings associated with the survey respondents. For example, are certain offices (or types of offices) overrepresented in that they required responses from their staff? Did certain offices highly discourage survey responses? Did the number of committee respondents respond in the same proportion as to their numbers in Congress, or are they underrepresented? Notably missing from the Speaker’s report are simple crosstabs by party, which could show that the parties differ greatly in key aspects, such as the gender and race of staffers promoted to leadership positions.
Finally, the report uses responses from House Officer Offices, but is very skimpy on the details of where those responses came from (which exact officer offices) or the weights from these used in measuring averages across the chamber. Both member and committee offices received full breakdowns by position in the full report, but the officer offices don’t receive similar treatments. This information is needed to verify how generalizable the findings are as they are filled with typically higher paying jobs with longer tenures.