PurposeThe purpose of this paper is to test the relationship between gross domestic product (GDP) and agency work.Design/methodology/approachThe paper develops a theoretical model for the time interdependence of GDP, agency work and regular employment and tested model predictions using a VAR model.FindingsResults show that on the macro level temporary agency work leads GDP development. Temporary agency work is an excellent instrument for employers to adjust the size of their workforce to fluctuations in product demand. Temporary work agencies, however, have a tough job finding qualified personnel in tight labour markets because workers generally prefer the security of a permanent contract. It is shown in this paper that, as a result of these two countervailing forces, the number of hours worked through temporary work agencies precedes GDP development. Agency work increases in the last phase of a recession after regular workers have been dismissed. It expands further, in line with GDP, when the trough is passed until agency worker's labour supply stagnates. This leads to a decrease in agency hours even before the business cycle reaches its peak. Then agency work declines further, in line with GDP, until regular workers are dismissed and the cycle start again.Originality/valueTemporary work arrangements have become a key area of interest for firms, academics and policy makers. This paper shows how the use of these work arrangement fluctuates over time. Also, this paper shows that agency work can be used in predicting future GDP development.