Comparing this year's hiring with previous years sounds simple and is full of traps. Getting it right is a genuinely useful analytical exercise, and the method transfers to any business comparison.
First: which measure are you comparing?
Four different series get described loosely as "hiring", and they can move in opposite directions in the same month:
- Vacancies / job openings - positions advertised. Measures intent.
- Hires - people actually started. Measures commitment.
- Net employment change - hires minus separations. Can be negative while hiring continues.
- The hires rate - hires as a percentage of employment. The right measure for comparing across years, because it adjusts for the economy having grown.
Comparing a count across years when the workforce has grown is the most common error. Always use rates for cross-year comparison.
The three base-year traps
Trap 1: the distorted baseline. Recent years contain extraordinary events - the pandemic collapse, then a sharp rebound with unprecedented vacancy levels. Comparing today against the peak of that rebound makes any market look catastrophic; comparing against the trough makes it look excellent. Neither comparison is informative.
The fix: compare against a pre-distortion baseline as well as the recent past. For most labour market series, 2019 remains the most useful "normal" reference point, alongside a longer ten-year average.
Trap 2: cherry-picked endpoints. Pick the right start month and you can demonstrate almost anything. The fix: show the whole series, not two points, and state your comparison convention before you look.
Trap 3: ignoring seasonality. Hiring has a strong annual pattern. Comparing December with September tells you about Christmas, not about the market. The fix: use seasonally adjusted series (statistical agencies publish them) or compare the same month year on year.
The method that works
- Get the hires rate and vacancy rate, seasonally adjusted, from the ONS (UK) or JOLTS (US)
- Plot at least ten years, so the recent distortions are visible in context
- Mark your comparison points: latest, one year ago, and 2019
- Look at the trend over the last six to twelve months, not the latest reading
- Repeat for your own industry, because sector divergence usually exceeds the aggregate move
What you will typically find
Two things, in most recent analyses of developed labour markets: that the aggregate figures are less dramatic than the discourse suggests, and that particular segments - entry level, specific sectors, specific regions - are much worse than the aggregate. Both findings are useful, and only the second one affects your search.
The general lesson
"Compared with what?" is the most valuable question in analysis, and the answer needs to be defended rather than assumed. A comparison against an abnormal baseline is worse than no comparison, because it produces confident wrong conclusions. See how hiring rates compare with five years ago for exactly this problem in detail.
Making fair comparisons is a core skill in the Statistics for Business course at London School of Business UK. Enquire today.