Independent work: choice, necessity, and the gig economy

Estimates of US independent workforce as % of working-age population
1 Includes those who work full time (16.9 million), part time (12.5 million), or occasionally (10.5 million) as consultants, freelancers, on contract, or performing 
temporary or on-call work each week.
2 Includes labor services and some leasing but excludes selling goods.
27
22
20
18
16
12
11
10
4
4
1
Intuit
MGI Survey
JPMorgan Chase Institute
Katz & Krueger
Katz & Krueger
<1
BLS
MGI Survey
MBO Partners
1
Burson-Marsteller
2
Kelly Services
Freelancers Union
MGI Survey
Online 
platforms 
only
Primary 
workers only
Primary and 
supplemental
workers
DUPLICATE from Report

---

103
McKinsey Global Institute
Independent work: Choice, necessity, and the gig economy
3. MGI SURVEY METHODOLOGY
In addition to drawing on the data and studies described above, MGI conducted a custom 
survey in collaboration with ResearchNow. It was administered to a sample of working-age 
respondents across the United States and five European countries (the United Kingdom, 
Germany, France, Sweden, and Spain). To arrive at the final statistics we report in this 
research, the responses for each of our European survey countries were population-
weighted and extrapolated to the full EU-15. 
The participants were carefully chosen as a representative sample of the population in each 
country controlling for age, gender, and geographic region. In the United States, we also 
controlled for household income, ethnicity and race, and Hispanic origin. The distribution 
of responses we eventually collected fell within 3 percentage points of the sampling targets 
for all control variables. Since we did not explicitly control for education, however, our final 
sample is somewhat more educated than the population at large. 
The survey was administered electronically in June and July 2016 and yielded a total of 
8,131 responses across all six surveyed countries, with at least 1,200 responses in each 
individual country.
It is important to note that not all analyses described in this report use the full number of 
responses, however. In particular, when describing the relative sizes of our four segments 
of independent workers, we only consider the subset of the population that engages in 
independent work. This results in a sample size of 2,259 independent workers across our 
six countries (as opposed to the 8,131 total responses). In the United States, the sample size 
of independent workers was 487 out of the 1,804 responses collected. Nevertheless, we 
did ensure that all results mentioned in the text of the report are statistically significant at the 
95 percent confidence level given the particular sample sizes in each comparison. Results 
that fell short of statistical significance were either explicitly reported as such or not included 
at all.
We acknowledge that our survey may not reflect the full extent of independent work that 
occurs in the informal economy. Individuals who are offline, those who have language 
barriers, and those who are paid off the books or do not have official immigration status are 
some of the most vulnerable independent workers—but by its nature, this activity is difficult 
to measure through surveys. 
Rather than asking independent workers to self-identify, the MGI survey asked each 
respondent to describe various characteristics of each of their income streams in order to 
be comprehensive and uniform across geographies. 
We designed the survey to capture each respondent’s complete work-life context (including 
all sources of active income over the past 12 months) as well as their satisfaction with their 
current state and their future desires. The majority of questions asked the respondents to 
select an answer from a pre-defined set of options presented in a randomized order so as to 
avoid introducing a bias. The core of the survey focused on the details of the jobs or income-
earning activities in which each respondent participated. It included questions focusing on 
flexibility, exclusivity, compensation, and duration of the work relationship. The full set of 
these questions was repeated for each activity to allow us to evaluate them at an individual 
level rather than focusing only on a person’s primary source of income. 
Since we asked respondents to describe their work situation over the previous 12 months 
rather than at a given point in time, we do not expect seasonality to have an impact on 
our estimates. While we translated the survey to the national language of each country 
and adapted it to the local context, the response options offered were designed to be 

---

104
McKinsey Global Institute
Appendix: Technical notes 
comparable across all the surveys. Exhibits A3 and A4 show the mapping used for key 
demographic variables in each individual country.  
Exhibit A3
SOURCE: McKinsey Global Institute analysis