"Job growth" figures are among the most reported and least understood economic statistics. Understanding how the number is constructed explains why it swings so much, why it gets revised, and how to read it properly.
What the number actually is
Monthly job growth is net employment change: total hires minus total separations across the economy.
Both of those gross flows are enormous - millions of people start and leave jobs every month in a large economy. Net job growth is the small residual difference between two very large numbers.
This has three immediate consequences.
Consequence 1: it is inherently volatile
When you subtract two large, independently varying quantities, the difference bounces around. A single month's figure carries a lot of noise, which is why economists look at three-month and six-month averages rather than the headline print.
Practical rule: never draw a conclusion from one month. Check whether the reading is outside the normal range of monthly variation before treating it as a change in direction - the same test you would apply to any business metric.
Consequence 2: it is heavily revised
Initial estimates come from incomplete survey returns and are revised as more data arrives - sometimes substantially, occasionally reversing the story. Statistical agencies are transparent about this; news coverage rarely is.
Practical rule: treat the latest month as provisional and look at where the revised series has settled.
Consequence 3: slowing growth is not job losses
This is the most common misreading. If net growth falls from 200,000 to 50,000 a month, employment is still rising - just more slowly. Job losses require the net figure to be negative. A "sharp slowdown in job growth" headline is frequently describing continued expansion.
What to look at instead of the headline
- The hires rate, not the net figure - it measures actual hiring appetite directly
- The quits rate, which reveals worker confidence and drives the vacancy chain
- Vacancies by industry, which turn before employment does
- Redundancies, which tell you about the separations side
- Employment level and rate, alongside economic inactivity - see what the headline unemployment rate misses
Together these tell you whether a slowdown means less hiring, more losses, or people leaving the workforce entirely - three quite different situations with the same headline.
Where to get it
The ONS Labour market overview for the UK, published monthly with commentary and clearly flagged revisions. The US equivalent is the BLS Employment Situation release plus JOLTS. Read the bulletins rather than coverage of them; the agencies explain their own uncertainty in a way headlines strip out.
Why the construction matters
This is a good example of a general principle: a statistic's construction determines how you should read it. A small net difference between two large gross flows is necessarily volatile and revision-prone. Knowing that is what stops you over-reacting to a single print.
The Statistics for Business course at London School of Business UK teaches exactly this kind of critical reading of published figures. Enquire today.