After how much time would the nonlinear effects of the subsidy be expected to be apparent? What hypothetical effect would discontinuation of the subsidy have:

After how much time would the nonlinear effects of the subsidy be expected to be apparent? What hypothetical effect would discontinuation of the subsidy have: while it's initially resulting in labor shortage reduction and then after time t? Are counterfactuals helpful for this problem?