#WhereDidTheJobsGo? – Health care and government most stable fields

So continuing our series of using excel voodoo (aka VLOOKUP) to take a look at where jobs the jobs went, let’s take a look at what jobs losses looked like across industries.

Unfortunately the data to break things down by Bermudian vs. non-Bermudian isn’t available so we’ll just look at jobs overall

How about construction?

Clearly this stands out as the industry that faced the most pain of the recession.  We overheated the economy and ultimately construction demand dried up causing the very painful loss of nearly 2000 jobs.

International business and associated business services have also had a painful slide with slight indications of recovery in 2014-2015.

It is too bad we don’t have a picture of pre-2008 due to the change in the occupational categories as it might give us more insight into what things looked like before and afterwards.

Then we can see retail, hotels and transport were all hard hit, likely especially due to the recession.  Restaurants seemed to hold steady.

Let’s move on to the growth sectors of the last few years.

Education, Healthcare and Social work was one of the few places that saw job growth. Perhaps unsurprisingly as Bermuda was increasingly encumbered with unemployed people needing assistance.

Oh, and government…

 

#WhereDidTheJobsGo – Comparing declines in major occupations by Bermudian status

Ah, so much to write about, so little time.  I haven’t followed up on our #WhereDidTheJobsGo analysis in a bit so I thought I’d throw up some more charts I’ve generated.

One thought I wanted to explore is what job declines looked like for Bermudians vs. non-Bermudians (excluding spouses and PRCs) on a relative basis.  \

Basically, if I zero the y-axis, plot Bermudians on the left, non-Bermudians on the right and adjust the upper values so that they start at the same spot, what do the declines look like relative to each other?

But what does it all mean?

Well, so far we’re seeing a story that non-Bermudian jobs on a relative basis declined first and more aggressively than Bermudian jobs across all job categories except Elementary Occupations.

It adds more questions to dig into.  Which jobs within major categories declined the most?  Why did Elementary Occupations see an increase rather than a decline?  Why did Plant and Machine Operators and Assemblers witness a spike in 2015?

Lots of questions, few answers as of yet.
What have we reviewed so far in this series?

#WhereDidTheJobsGo? – taking a look at cumulative changes in job numbers

Continuing on my analysis of the Employment Report statistics I thought I’d take a look at cumulative changes in jobs by major occupation and see what that looks like.

This gives a much clearer of the absolute number of jobs lost by Bermudians in the last few years.  For example, Clerks (aka. Secretarial positions) suffered a substantial decline from 2011.

An interesting point of note is the spike in Service worker jobs from 2011 – 2013 along with a minor spike in professional jobs which seemed to buck the trend.

The question is what caused this spike? My current hypothesis is that this is related to the hospital project but we’ll likely get a better idea if we dig further into the breakdowns of the jobs within individual categories.

If we look at the same data but for non-Bermudians (excluding spouses and PRCs) we see quite a different picture

There were more aggressive declines and larger numbers of overall jobs lost in the Professional, Service worker and Trades worker categories.  Notably there was no upwards spike for 2011 to 2013, only a slight change in rate of decline in a few occupational types.  Also interesting is the switch from heavy decline to an uptick right around 2014 which coincides with the ending of the term limits.

Interesting…

Note: the charts are based upon summing the cumulative yoy changes in job numbers

 

Where did the jobs go? – taking a look at major occupational groups

So far in our attempt to answer the question of “Where did the jobs go?” we’ve seen that since 2008 a significant number of both Bermudian and non-Bermudian jobs have declined. Let’s start digging a bit deeper to see what the data can tell us.

Of note, the data from the Employment Survey Tables provided by the Department Of Statistics unfortunately had a change in occupational category groupings from pre-2008 to post so I’m only going to show data from 2008 for now.

Also note, only Bermudians and non-Bermudians (not including Spouses of Bermudians and Permanent Residents) are being considered here.  Simply because the the other numbers don’t vary enough to warrant detailed analysis at this time.

So, let’s take a moment to examine what happened with Bermudian jobs when grouped by major occupation.

This tells an interesting story of which Bermudian job occupational types were most impacted by the job losses over the last 8 years.  Notably jobs in the “Clerk” role (these are effectively Secretarial positions and the like) were heavily impacted.  Most except service workers experienced declines.  Technicians and agriculture workers seem to have been un-impacted. It is interesting to note that the declines are fairly steady with a bit of an acceleration in the job losses in the clerk role from 2011.  Thankfully from 2014 to 2015 most roles stabilized.

As a contrast, let’s take a look at non-Bermudians (not including Spouses and PRCs).

For non-Bermudians, most roles experienced heavy declines.  Professionals were the largest in losses with nearly 1/3 of roles disappearing.  Service workers and trades worker jobs were also very heavily impacted.  Notably there has been an ever so slight uptick in a few of the roles.

What is the cause of these declines?  It is hard to say at this point. What we do know is that the term limit policy was introduced in 2007 and ended in 2013 and that could have had an impact. In order to get a better idea we’d likely have to review work permit statistics in depth, if such information is even available at a granular level. There was also a global recession during that time period. Clearly was is obvious from these graphs is that non-Bermudians roles were eliminated earlier and significantly faster than Bermudian roles when looking on a basis relative to the total numbers of each.

 

Where did the jobs go? – Examining the rate of growth vs decline

Continuing our investigation into the question of where the jobs went I’ve done some additional analysis on the rate of growth vs. decline in Bermudian vs. non-Bermudian + Permanent Residents (PRC).

A reader raised the question of whether PRCs changing status might skew the numbers so I’ve included them as part of the “non-Bermudian” numbers

Let’s start by comparing overall Bermudian vs. non-Bermudian numbers since 2003.  In this chart I’ve provided illustrations of both non-Bermudians with and without PRC’s being included.  It is worthy to note that from the numbers it looks like most PRC job growth occured leading up to 2009 and then stabilized in a range between 807-842.

It is important to note that the chart is split with Bermudian numbers guided by the numbers on the left axis and non-Bermudians on the right axis.  This allows a for better visual relative comparison, though importantly both axis are zeroed so as to give a fair relative comparison.

What we see here is a confirmation of yesterday’s analysis where Bermudian jobs peaked earlier and have been in steady decline while non-Bermudian jobs had a sharp peak in 2008 and had a more aggressive decline.

Let’s look at that from another perspective isolating just the declines from 2009.

Again note, Bermudian numbers are relative to the left axis, non-Bermudians to the right and both axis are zeroed so as to not manipulate the perception of the numbers.

This chart gives a better perspective on the relative decline of non-Bermudian + PRC jobs vs. Bermudian.  What it shows is that when considering the decline in jobs by status vs the total number of jobs by status, the decline in non-Bermudian jobs was far more severe than Bermudian.

So why does it matter to compare the relative declines of non-Bermudian vs. Bermudian jobs?  Well it is the belief of this writer that imported highly-skilled highly-compensated labour is a major driver of Bermudian job growth and declines.  This is a hypothesis, not a fact so we need to examine the numbers to see if there is a correlation that would justify that belief.

There are roughly 3 times as many Bermudian jobs as non-Bermudian jobs.

The ratio itself hovered in a band between 2.5 and 3.2.  It is the suspicion of this writer that the only reason why we had a dip in this ratio from 2005 in that Bermudians were near full employment.  Verifying that will require a request for the raw data from the Labour Force Surveys going back to 2003.  That will be a question we’ll have to come back to when I’ve gone through much of the data the Statistics department has already provided.

None the less, generally the question is whether there is a correlation between non-Bermudian jobs and Bermudian jobs.  Generally speaking, does every non-Bermudian job create 3 Bermudian jobs?  If this is true and we want full Bermudian employment, how many non-Bermudian jobs do we need? If we have an oversupply of non-Bermudians, what impact and repercussions does that have and how do we prevent it?

 

Where did the jobs go?

One of the big questions I’ve been wanting to be able to answer is “where did the jobs go”?  It is rather easy to guess where they went but guessing isn’t very informative. It is far more useful to have survey numbers to do some analysis.

Thanks to the Department of Statistics who have fulfilled my request for raw Employment Survey Tables from 2009-2015, I can do some long overdue analysis of job trends.  (Which is far more interesting than political commentary)

As a teaser, here’s a couple charts taking a look at how employment numbers have changed since 2003. (Why 2003?  Simply because that’s what data I have available thanks to prior requests)

So let’s start by taking a look at Total Filled Jobs by Bermudian Status.

This chart gives us a big picture view of what has happened employment wise in Bermuda over the last 10+ years.

Some points of note, Non-Bermudians (Other) means non-Bermudians not including spouses and permanent residents, who are listed separately.  Also, Permanent Resident numbers weren’t available in 2003

The big takeaways here is that

  • there was a significant decline in total jobs from the 2008 peak of 40,213 to a hopeful bottom of 33,319 in 2015, a decline of 17%.
  • Bermudian jobs peaked in 2004 at 27,443 and have a hopeful bottom in 2015 of 23,576, a decline of 14%
  • Non-Bermudian jobs (not considering spouses and permanent residents) declined from a 2008 peak of 10,367 to a 2014 bottom of 6,885, a decline of 33.5%
  • Spouse of Bermudian jobs remained fairly stable, fluctuating near 1,950
  • Permanent Resident jobs grew from 286 in 2004 or 830 in 2015

Here’s a different perspective of the same numbers as a stack chart.

The 2004 peak of Bermudian jobs suggest that this may have been due to Bermudians reaching full employment.  In other words, there were more job opportunities than there were Bermudians willing to fill them.

One question that needs to be raised here is that if this indeed was the case, why did we massively stimulate the economy with a variety of capital projects financed by a ton of extra debt between 2004-2008?  All this when it is unclear what we stood to gain if we already had full employment?

Full Bermudian employment should be our goal.  We reached it and kept our foot heavily on the gas pedal of our economy.  That is, until we slammed on the brakes by introducing term limits at the same time as a global recession.  What impact did that have?  Hopefully the numbers will tell us more.