Thursday, January 16, 2014

Teen Employment and the Minimum Wage, 60 years of experience

(Updated with trendlines at bottom chart, and better trendlines on top charts).
(Updated again to show the bottom chart in Employment to Population Ratio terms instead of total employment)

Added: For those visiting from Matthew Yglesias' Slate.com post, (1) I use teen employment because the higher % of teens at MW makes the trends show up more clearly, but I find the same trends in the total employment numbers, and (2) it's important to consider the difference between an event-based analysis and long-term trends.

And, for everyone, here is an updated post that adds recessions to the graphs. And a post with a regression of teen EPR against RGDP and the scale of MW hikes.
And here is my last post on the topic for now.


Here are the teen employment trends during all 7 of the series of minimum wage hikes (I exclude the high inflation period in the late 1970's, which included several nominal hikes that roughly matched inflation).



In every episode, except 1996 (which is the smallest hike relative to average wages), there was a distinct decline in the trend of teen employment, over the period of time covering from a few months before the initial hike until a few months after the follow-up hike.

Here is the entire period:
Pre-trend lines are for period from 27 to 3 months before MW hike.
MW Trend lines are for period from 3 month before to 27 months after initial MW hike.
[Y axis = Total teen employment (000's)]
 
ADDED:  Here is the graph in terms of the Employment to Population Ratio, which is probably a better way to do it.  The basic story is the same, though.  After 30 months, EPR, on average, is about 6% lower than it would have been based on the previous trend.  Using EPR does help to see that the scale of the effect is probably inflated by double counting, because the positive trends coming into some MW episodes probably reflect some amount of rebounding from the previous shock to teen employment.
Pre-trend lines are for period from 27 to 3 months before MW hike.
MW Trend lines are for period from 3 month before to 27 months after initial MW hike.
[Y axis = Teen Employment To Population Ratio]

Every kink down in the trend comes in the few months leading up to a minimum wage hike, except for a brief decline in 1969 and the contraction beginning in 2000. (The period from 1976 to 1982 is the period where frequent minimum wage hikes and high inflation rates make it hard to distinguish between categories.)

Is there any other issue where the data conforms so strongly to basic economic intuition, and yet is widely written off as a coincidence?

15 comments:

  1. Out of interest, what determines the end of a "minimum wage hike" in your final graph? Is the beginning of a white section of graph mark where wages are returned to "pre-hike" levels? Or is it something else?

    ReplyDelete
    Replies
    1. Since MW hikes tend to come in series, I have been analyzing it by treating the 2 years after the first hike as a single period of time. It's somewhat arbitrary. If you click on the "Minimum Wage" tag, you'll see some other posts that go into more detail about how I've been looking at the data. The red "MW Trends" correspond to the shaded sections plus 3 months on each end.

      Delete
  2. Anon: I assume shaded refers to years in which the nominal federal MW changed..

    See: http://www.dol.gov/whd/minwage/chart.htm#sthash.hBnO1zUt.dpuf

    ReplyDelete
  3. There's an explanation for the atypical trends you've observed around the year 2000 - the inflation and deflation phases of the Dot Com Bubble between 1997 and 2003 really skewed the demand curve for employing young Americans during those years. Later this week, we'll show that the apparent aberration throughout that period is directly proportionate with U.S. stock prices at that time.

    ReplyDelete
  4. Certainly very suggestive, but what exact measure of teen employment are you using? Raw numbers? Percentages? Shouldn't you be normalizing this somehow, such as by total number of employed people to make sure you aren't being mislead by other, more general trends in the economy? At the very least, shouldn't you have highlighted periods of booms and busts? I mean, we'd want to know the relative importance of the minimum wage compared to these other factors, particular growth (in GDP and employment)...

    Don't get me wrong, this really is very suggestive and interesting, it's just lacking a few more controls to make sure that you've really shown what you think it's shown.

    ReplyDelete
    Replies
    1. It's just raw numbers. By treating each series of MW hikes as a single incident, I can divide time into distinct periods that either saw (1) declining real MW levels or (2) were marked by distinct positive shocks in MW levels. I leave out 1976-1982 because this period was marked by multiple MW hikes and high inflation, so that the time periods do not fit into either of the two distinct categories. I have been defining these MW periods as 2 years long, which is kind of arbitrary.

      This leaves us with 7 episodes. So the problem, statistically, is that this leaves a very small number of distinct observations, and as you noticed, the underlying time series is marked with a number of incongruities that would be hard to normalize well.

      So, in the end, looking at this from a top-down perspective, in this way, we almost have to look at it anecdotally.

      What is surprising to me is how, with this stair-step pattern throughout the time series, except for the 2000 contraction, all the steps down occur strikingly within a few months of one of these 7 episodes.

      Delete
    2. Thanks for the reply! And I take your point, short of actually trying to do some sort of regression analysis with appropriate controls (a substantial challenge that would be difficult to pull off well), this is suggestive and striking evidence, if still tentative, that the minimum wage seems to have a direct effect on teenage employment trends.

      Very interesting, my only further recommendation would be to actually label the graphs indicating that it's the total number of teens employed.

      Delete
  5. You might consider correcting for the general level of unemployment. For instance, regressing the teen unemployment rate on U3 and then doing the same analysis with the residuals. Alternately, you could extend the analysis to black unemployment and see if there is a relationship there.

    ReplyDelete
    Replies
    1. I did look at this, and as anonymous points out below, employment tends to be a better indicator, because MW workers seem to have a higher propensity for dropping from the labor force. That makes sense, because they are usually part time, marginal jobs, not held by heads of households. Using UE or looking at the entire population, the same trends tend to hold, but with less significance.

      The relationship regarding black unemployment, and especially black teen employment is very strong. However, the data I have found only goes back to the early 1970's, and because I am treating each series of MW hikes as a single observation, that only leaves 4 episodes, which makes it even harder to attach any significance to the relationship.

      Delete
  6. Hikes in the minimum wage tend to be relatively small (compared to the economy as a whole) and so end up primarily effecting marginal workers (this seems very obvious to me) which generally means knocking them out of the labor force, which drops them from the unemployment rolls, and makes the effect invisible to those using UE to measure the impact of the increase in the MW.

    ReplyDelete
  7. Some comments at marginalrevolution caused me to look more closely at the timing of MW hikes to recessions:

    I decided to look at it in more detail:
    03/1956: UE=4%, recession starts in late 1957

    09/1961: UE>6% but recovering, recession had ended in early 1961

    02/1967: UE<4%, no recessions

    05/1974: UE=5%, recession started at end of 1973, unemployment shot up in 1974

    04/1990: UE~5%, recession started soon after and unemployment shot up 1990

    10/1996: UE~5%, no recession

    07/2007: UE<5%, recession started soon after and unemployment shot up in 2008

    So, 3 of the 7 were enacted just before a recession, one at the beginning of a recession but before unemployment had shot up, and three unrelated to recessions. So, as a precursor to recessions, it's 3 or 4 out of 7. Looking at this closer, while there is clearly some significant effect on employment, depending on whether the post-MW context is recessional or not, the timing of the recessions cannot explain why there is a consistent downward kink in employment that begins in the few months before the MW hike. For recessions to explain this pattern, MW hikes would have to come on the heals of recessions. 1974 is the only episode that fits that timing. - See more at: http://marginalrevolution.com/marginalrevolution/2014/01/teen-employment-and-the-minimum-wage-sixty-years-of-experience.html#sthash.bDXj4hkH.dpuf

    ReplyDelete
  8. realclearpolitics.com posted a link to my post:
    http://www.realclearpolicy.com/blog/2014/01/21/does_the_minimum_wage_cause_recessions_807-comments.html#disqus_thread

    He brings up some questions I've been hearing from several sources, and I think I dealt with some of those issues in the background, without necessarily addressing them explicitly. So, I'm reposting my comment here, just to help add clarification to the conversation, in case anyone is reading the comments with the same questions in mind. Here's the comment:

    Thanks for looking at this. I have looked at the issues you brought up, so I have some thoughts about them.
    Teen employment tends to respond to both recessions and to minimum wage hikes in a proportion higher than total employment does. I'm not sure that the signature of that ratio is different enough between the two events to be able to delineate between the two sources of unemployment. The reaction of total employment to minimum wage hikes is similar to the reaction of teen employment, but to a smaller degree. Intuition suggests that as the minimum wage is increased, older workers would substitute for teens. But, the bls annual characteristics of min. wage employees over the past few years shows a surprisingly similar relative increase in the number of min. wage workers across demographic groups as the MW increased, so I'm not sure the intuition is accurate.
    On the timing of recessions, The 1974 hike did happen just as a recession was kicking into gear. 1956, 1991, and 2007 happened before the first signs of recession. 1961 happened coming out of a recession. 1967, and 1996 were not near recessions. So the interplay between MW hikes and recessions might be something interesting to think about, and the appearance of a recession after MW hikes would certainly tend to turn the employment trends down in those cases. But, the main thing that I thought was interesting about the data was that the downward trend in every case began in the few months leading up to the MW hike. The timing is as interesting as the scale here. And, I don't think the business cycle explains that.

    ReplyDelete
  9. "Then, again except for the contraction beginning in 2000, every period after minimum wage hikes is associated with rising employment, relative to trend."

    Don't you mean declining employment?

    ReplyDelete
    Replies
    1. Yes. That is awkward wording. I've edited it. Thanks.

      Delete
  10. Yes, education. Educational research promotes greater flexibility, self-directed learning, right brained learning, and holistic and overhauled grading policies. Of course, public education (and increasingly private education) is implementing the exact opposite.

    ReplyDelete