Not so black and white

This years employment brief has been released and will likely be a focus as I’m able to find time over the coming days to review it.

The Royal Gazette has already jumped on the earnings gap between black and white workers by opening their coverage with “The average salary of white workers is 40 percent higher than blacks, according to a recent employment survey.”  This reflects poor reporting because the implications of the differences in racial demographics of Bermudians vs. Non-Bermudians are not made clear.  Thus Bermudians could potentially be falsely led to believe that the racial earnings gap is a lot larger than it is when the sheer percentage of white non-Bermudians who are brought to the island massively tip the scale.  The Royal Gazette can do better than this.

It is also a failure on the part of the employment brief (just as we highlighted with past Cure Reporting) to not identify racial earnings based upon Bermudians vs. Non-Bermudians.  The reason being that the majority of Bermudians are black while the majority of non-Bermudians are white and yet the level and type of jobs held by Bermudians vs. non-Bermudians, especially by race, are quite different.  This is also the reason why Bermudians median income lies at $51,976 while non-Bermudian median income lies at $65,316.  Let us also note that this is median and not average which means the person in the middle of the group not the average salary taking the total and dividing by all members.  In the case of average salary the numbers would likely be even more skewed.

To get a better idea lets look at some hard numbers by taking a look at data available from the Annual Review of the Workforce Survey Report 2006 (used for comparison, it’s the latest one I have on hand and no others are online yet) and take a look at Gross Annual Income by Race and Bermudian Status, 2006.


Note that the sheer number of Black Bermudians skew the chart and outpace anything else in the lower 3 categories of pay ranges from less than $23,999 to $60,000 – $95,000.

Let’s again look at this in % comparison


Note how white non-Bermudians dominate the upper pay ranges while Black Bermudians dominate the lower pay ranges.

To make things clearer let us first compare Bermudians and then compare non-Bermudians alone.


So here we have Gross Annual Income By Race for Bermudians only.  This is where we need to focus our attention on discrepancies of where the percentage distribution does not match general demographics.  We cannot expect Bermudians to measure up equally against the best and brightest of the rest of the world and thus the fairest comparison is to compare Bermudians to Bermudians.

Now let’s look at non-Bermudians


Note how much white non-Bermudians skew the upper portions of the pay scale.  How can we honestly simply compare black and white when not taking Bermudian/non-Bermudian into consideration?

For those wondering why the earnings gap exists between the races we covered that back in November 2007.  As we concluded the earnings gap between the races statistically (there is still the potential for racism to be a factor as to why people holding the same qualifications make different amounts of money, though those kind of stats are not available to my knowledge) comes down to profession which subsequently comes down to education

This entry was posted in Uncategorized by . Bookmark the permalink.

7 thoughts on “Not so black and white

  1. Headline’s like that play into the hands of those who wish to use race to divide to country. There is something wrong with the distribution of wealth yes and that should be looked into of course but, to misuse the statistics at hand and therefore unintentionally (I hope) exaggerate the situation does nothing to help racial tensions.
    I’m glad you’ve taken the time to really take a look at the statistics. I was wondering myself whether Non Bermudians had been included and if that skewed the statistics and now I have my answer.

  2. The top level numbers tell us nothing except that we have an immigration policy which protects positions for Bermudians. The outcome of this is expats fill positions that Bermudians are unqualified or unwilling to perform.
    The unable means higher compensation to the expat. The unwilling means they go to low paid foreign workers from poorer countries. The high earning expats dramatically skew the average because we’re throwing in CEO’s of billion publicly traded corporations who make many millions a year.
    So these stats tell us nothing other than confirming the structure of our immigration laws.
    Comparisons by job position would be useful. But my understanding is that Statistics haven’t released those yet. Ironic, as those are the most insightul.
    Then we could really see any structural inequities by gender, nationality, age, race etc..

  3. These are good charts that tell a deeper story than the simplistic whites make 40% more headline.
    Also, I know there are valid reasons why median income is used instead of mean because of bloated salaries at the top, but isn’t it still a bit misleading? Especially since the Gazette switches quickly back and forth between the two sets of values.

  4. Actually I clarified median vs. average (mean) to allude to the Gazette’s usage.
    “While Bermudians rake in an average of $51,976 and Permanent Residents $56,769, non-Bermudians earn a median annual income of $65,316.”
    Unfortunately I don’t think the author of the article understands the difference between median and mean because in the actual employment brief mean is primarily used and I’m not sure average is ever provided.
    As Christian suggests the stats alone are not very useful unless they’re broken down by job position.
    Ideally we would see the stats broken down by job position, race, Bermudian status and education level (or better, specific degree/qualifications as I’ve noted not all bachelors are created equal). That would give us a true picture we could use to establish areas that need to be improved.
    I’m rather pleased to see people haven’t incorrectly jumped on these stats to rant about inequalities.
    I also am hoping to get some time to revisit my projections for a downturn in IB workers, which didn’t happen indicating term limits may not have had the impact many were expecting.
    Also due are cautions on the unfettered growth in construction workers and the impacts that may have on having overheated our construction industry with a bubble and the subsequent collapse that may result. In the future it is worthwhile studying the infrastructure needs of the community using statistics to judge approvals for work permits and building projects.

  5. I am not a statitician, but why do they use bands like:
    $24,000 – $59,000
    $60,000 – $95,000
    $96,000 – $155,999
    Greater than $156,000
    instead of
    $25,000 – $49,999
    $50,000 – $99,999
    $100,000 – $149,999
    Greater than $150,000
    Do they try and adjust the visual results by doing this, or is there some real rationale
    Just wondering.

  6. Pitts Bay,
    That’s actually my fault. I had to retype out everything by hand and in doing so wrote the numbers wrong on the charts. The right half of the numbers should end in 999.

Comments are closed.