Constraint and opportunity: identifying voluntary non-employment

This paper is an attempt to assess the extent to which the behaviour of an individual is the result of the constraints that he or she faces – factors beyond individual control - or the result of the exercise of his or her preferences. The study concentrates on participation or non-participation in employment, with non-participation defined broadly to include fulltime education, caring or early retirement, as well as unemployment. Following a discussion of potential methodological difficulties, data from the British Household Panel Study are used to construct models of the probability of being in employment, controlling for various constraints. Starting from the position that all non-employment is voluntary, possible constraints are introduced in layers corresponding to the degree to which they are regarded as beyond individual control. The layered approach allows for the fact that opinions vary as to what factors it is appropriate to regard as constraints. Predicted probabilities of being in employment are then compared to each individual’s actual state. If the model predicts that he or she has a high probability of being in work, and in fact he or she is not, then there is a prima facie case that she or he is voluntarily out of work. However, since there may be unobserved constraints, the outcome is cross-checked by starting from the opposite position, namely that all non-employment is involuntary, then gradually subtracting those for whom there is evidence of having chosen to be out of work. Only those who are found not to face significant constraints and who state that they do not want work can with confidence be asserted to be voluntarily non-employed. The results suggest that after taking into account as many constraints as possible, one-tenth of the non-employment in our sample is unambiguously voluntary, with a further one-tenth being indeterminate.