Direct payments are regarded as a suitable instrument to safeguard jobs in the agricultural sector. However, empirical findings to date do not unambiguously support this expectation. We further empirically investigate this research question on dairy farms with a focus on family work. Using a spatial regression discontinuity design, we rely upon selection on unobservables assumptions. The Swiss direct payments system creates a discontinuous jump near the border of agricultural production zones for the amount of public subsidies a farm receives. Using two-stage least squares (TSLS) to estimate the policy-relevant effect, we find that an additional CHF 50,000 can generate a job for a female family worker in the dairy sector. Male employment is not affected. These results show that direct payments can safeguard traditional family farming.