The Law of Demand and the minimum wage: It applies to number of hours worked, not the level of employment

From Mark Perry.

"

Posts on CD about the minimum wage always generate a higher than average number of comments, and Friday’s CD post (“Do Demand Curves Slope Down or Not?“) was no exception – there have been 46 comments so far. Most of the minimum wage debate centers on the issue of whether minimum wage increases have any effects on employment levels. Specifically, does the empirical evidence point to any significantly negative effects on employment levels following minimum wage hikes, as clearly predicted by economy theory? Some empirical evidence like the much-cited 1994 study by Card and Krueger found “no indication that the rise in the minimum wage reduced employment” at fast-food restaurants in New Jersey following a minimum wage increase to $5.05 per hour compared to nearby fast-food restaurants in Pennsylvania where the minimum wage remained constant at $4.25.

Let me attempt to reconcile the apparent inconsistency between: a) economic theory, which clearly predicts a negative relationship between the minimum wage and the quantity of unskilled workers demanded by employers, and b) some of the empirical evidence that finds no negative employment effects following minimum wage hikes. Here’s the key point: The negative relationship predicted by economic theory is not: a) between minimum wage hikes and the number of unskilled workers employed, but b) between minimum wage hikes and the number of unskilled work hours demanded by employers. 

The two charts above help to illustrate that difference:

In the top chart, we see a negative relationship between an increase in the minimum wage and the number of hours of unskilled work demanded by employers in the 12-month period following the increase in the hourly price of unskilled labor (to capture the effects on future hiring). Like an increase in the cost of any other labor input or other input like food, energy, raw materials, machinery, equipment rental, or building rent, employers facing a 39% increase in the cost of unskilled labor (from $7.25 to $10.10 an hour) would have no other choice than to reduce the number of unskilled work hours – it would simply be a necessary strategy for survival. As I pointed out recently, a minimum wage increase to $10.10 per hour would be the equivalent to an annual tax of more than $6,000 per full-time worker earning the minimum wage.

The various strategies employers might use to reduce their demand for unskilled work hours over the 12-month period following a 39% minimum wage hike might include: a) reducing the number of hours worked per week by entry-level unskilled workers, e.g. cutting their hours from 40 to 30 per week, or from 30 to 20, etc., b) reducing the number of unskilled workers currently employed through layoffs, c) reducing the number of unskilled workers that employers might have previously been planning on adding to staffing levels in the future, d) substituting skilled workers for the now relatively more expensive unskilled workers, and e) investing in technologies that would substitute automation, mechanization, robotics, and self-serve options for unskilled workers. Although the effect of a 39% minimum wage hike on employment levels might be uncertain, the negative effect on the number of hours of unskilled labor demanded by employers would be much more certain and predictable according to the Law of Demand. 

The bottom chart shows graphically how it would be possible that an increase in the minimum wage might not adversely affect the number of unskilled workers employed by looking at the relationship between the average weekly compensation for unskilled workers (and not the hourly monetary wage) and the number of unskilled workers.

Suppose that employers realistically respond to a 39% increase in the minimum wage by: a) cutting hours for minimum wage employees and b) reducing or eliminating  non-monetary forms of compensation that might include free or reduced cost food, merchandise, uniforms or parking, bonuses/profit-sharing, educational reimbursement, paid holidays and company parties/picnics, health care benefits, etc. Following a reduction in hours and non-monetary benefits, the average compensation per minimum wage employee might be unchanged, as could be the number of workers employed at the minimum wage. That is, if the employer can completely offset the monetary increase in the minimum wage by a reduction in hours and fringe benefits, there will be no need to reduce overall staffing levels. In that case, empirical evidence may find no negative relationship between increases in the minimum wage and employment levels, even when there’s a predictable and negative effect on the number of hours of unskilled work demanded by employers.
Bottom Line: It’s more accurate to say that the Law of Demand predicts: a) a negative relationship between higher wages and the number of hours of unskilled work demanded by employers, rather than b) a negative relationship between higher wages and the number of unskilled workers employed. Therefore, it’s possible that a minimum wage hike won’t always negatively affect employment levels for entry-level unskilled workers, but will affect the number of hours demanded by employers for unskilled labor. That’s how we can reconcile the apparent inconsistency between economic theory and some of the empirical evidence…..

Update: Here’s another thought. Some studies find no significant employment effects when comparing employment levels of unskilled workers before and after an increase in the minimum wage. But that doesn’t capture the possible increases in the number of unskilled workers employed that might have taken place without an increase like 39% in the minimum wage from $7.25 to $10.10. For example, suppose that employers had been planning to expand their operations over the next several years and had planned to increase staffing levels of unskilled workers by 1-5%.

Following the minimum wage hike, they cancel plans to expand their operations and instead maintain current staffing levels by reducing hours and benefits. Therefore, we might find that a 39% increase in the price of hiring unskilled workers had no significantly negative effect on current employment levels, even when there could be a significantly negative effect on the number of unskilled workers hired in the future! The 1-5% planned future increase in hiring unskilled workers never happens; but those jobs that weren’t created aren’t visible and aren’t accounted for when we try to calculate the negative effects of the minimum wage."