Time Matters

Having spent a lot of time in the past few weeks looking over minimum wage studies, surveys, and summaries, I noticed something important to note: the studies are consistent with the Second Law of Demand.  The Second Law of Demand states that the longer a price change persists, the more elastic the demand curve becomes.  This is very true in discussing minimum wage.  The studies that show little/no negative employment effects tend to be short-term studies, whereas the ones with more pronounced negative effects are longer-term.  In fact, in one of the longest outlooks, GMU economist Walter Williams looked at teenage unemployment and labor force participation rates among blacks and whites and he finds dangerously negative consequences (he finds in 1948 that white and black teenage unemployment rates were about equal and black LFPR was higher than whites.  Now, black teen unemployment is around 28% (has been as high as 50%), white teen unemployment is around 13%.  Since 1948, black teen unemployment has averaged about 19.5 percentage points higher than  white teen rates.  Williams attributes this to the minimum wage (as well as other factors).

The lesson here is that time does matter.  The longer a price control is left in place (or, God forbid, it be indexed to inflation), the more pronounced the negative effects will be.  In the short run, there may be little/no effects (a la Card-Kruger), or the effects may be negative, (a la Mark Perry), but there’s really too little information to tell.*  There’s not much a person can do in the short run to adjust.  In the longer run, however, his ability to adjust is much greater.  In the case of the minimum wage, a person can bring in capital (or invent new equipment) to offset the increases.  He can retool/reconfigure his workforce.  In the longer run, the margins he can adjust along are more numerous, allowing for more of the negative effects of minimum wage to appear.

This also has consequences on cost-benefit analysis.  Although I have made it clear I do not believe any meaningful cost-benefit analysis can be done at the macro-level, there are those who do and thus bravely plow ahead.  But, given the longer-run issues discussed above, the cost-benefit analysis would need to take into account an ever-elasticizing demand curve.  The costs would continue to rise at a quickening pace the longer the trend persists (the exact nature of that trend I will leave to some econometrican to figure out), eventually surpassing the benefits.  This would need to be taken into account when doing any kind of analysis or having the conversation.**

Time matters because economies are dynamic, not static.  Failure to take into account this simple fact will lead to poor decision-making.

*None of this is to say short-run analysis doesn’t have use, of course.  What it is to say is to urge caution from drawing conclusions based on short-run data.

**Perhaps some math-savvy person can figure out where exactly this inversion occurs.  it won’t be me, however, as I find minimum wage incredibly immoral and  that arguing sacrificing some people is ok so long as another group benefits more is outright despicable.  I fear if I were to develop such a model, it’d be used for determining the “exactly right” minimum wage and I will not have my name tied to such a thing.

4 thoughts on “Time Matters

  1. I agree that time matters but I wonder if it matters in a different way than you’ve put forth.

    For example, let’s conjecture that minimum wage laws had a large impact in 1948 but with ever decreasing impact between then and now, and that now they have minimal impact.

    How would the data look different?


      • It’s hard to make clear.

        When Card & Krueger came out all those years ago, I was quickly looking through various minimum wage studies to see how they compared and it looked to me like the older the study, on average, the largest adverse impact raising the minimum wage had on employment. So my first question is: did you notice a similar trend?

        If your answer to that first question is no, then the rest of my question doesn’t make sense, so I’ll wait to see if you found any trend over time while review the studies.


Comments are closed.