McKinsey Counted 162 Million and the Caveats Started on Page Two

The Number and Its Footnotes

In October 2016, the McKinsey Global Institute published Independent Work: Choice, Necessity, and the Gig Economy, a report that placed 162 million people in independent work across the United States and the European Union. The figure landed in parliamentary briefings, Commission documents, and newspaper headlines within weeks. It remains one of the most-cited statistics in the entire gig-economy literature.

The report's authors arrived at 162 million through a deliberate act of aggregation. They drew on existing national surveys — the Bureau of Labor Statistics Current Population Survey for the US, and Eurostat's Labour Force Survey for the EU — then supplemented them with a purpose-built survey of roughly 8,000 respondents across six countries. That supplementary instrument asked not only about primary jobs but about secondary and occasional income-generating activity, which official labour surveys typically miss or undercount. Pulling in that secondary work was the decision that drove the headline figure upward.

The methodology section, beginning on page two of the technical appendix, named the problem openly. "Independent work" as the report defined it covered four distinct populations: free agents who chose independent work as their primary livelihood; casual earners who supplemented wages with occasional gigs; reluctant independents who wanted traditional employment but could not find it; and financially strapped workers doing supplementary gigs out of economic necessity. These categories behave differently in the labour market, carry different policy implications, and would not ordinarily be counted together in any official statistical framework. Aggregating them produced a large, attention-commanding number that was also, by the report's own admission, not directly comparable to any official measure.

The definitional reach also raised a comparability problem across borders. Eurostat's own-account worker category, which underpins EU self-employment counts, excludes employers — people who hire others — and excludes contributing family workers. McKinsey's framework included some of those adjacent populations. The result is that the 162 million figure is neither a BLS category nor a Eurostat category; it is a McKinsey category, constructed to answer a McKinsey research question.

None of this makes the report dishonest. The authors were transparent about their choices, and the four-quadrant typology remains analytically useful for distinguishing workers who chose independence from those pushed into it. But the gap between the headline and the appendix illustrates a persistent problem in independent-work statistics: the number that travels is rarely the number that was actually measured. Every count of independent workers is, in part, a count of the definition used to produce it.