Legislation and anticipation: Consequences for the measurement of the minimum wage employment effect

When Congress or a state legislature passes a minimum wage increase, there is often a period of several months between the passage of the increase and its implementation. During this period, employers know the level of the current binding minimum wage, but they also know the level of the minimum wage at a defined future date. Although this “future wage,” as I call it, is not yet binding, employers may take action to decrease their employment in anticipation of the coming increase. Using monthly CPS data from California, Oregon and Washington in the years 19942014, I estimate the effect of both the current minimum wage and the future wage, when it exists, on the teen employment-population ratio. I divide the observations into two regimes: a “limbo” period during which there is a future wage, and a “normal” period during which there is not. I find that while a strong correlation exists between the minimum wage and teen employment in the normal regime, this correlation disappears in the limbo regime when a future wage exists. This indicates that although employers make employment decisions based on the current wage when no future wage exists, it is not clear whether they continue to do so when a future wage is present.