Self-Professed Policy Analyst Predicts Minimum-Wage Induced Employment Disaster [CORRECTED]

[Graph corrected 3/18 of 3/15 post– apologies to all for the error of using NY-wide series in the earlier post] New York City. Inveterate commenter Steven Kopits cites an article from some publication called “Liberty Nation”:

76.50% of full service restaurant respondents reduced employee hours, and 36.30% eliminated jobs in 2018, in response to mandated wage increases. 75% of limited service restaurant respondents report that they will reduce employee hours, and 53.10% will eliminate jobs in 2019 as a result of mandated wage increases that took effect on December 31, 2018.

This is a pretty faithful summary of the report by the NYC Hospitality Alliance. I thought that instead of citing (unweighted) statistics, it might be useful to look at data, to see what actually happened (as opposed to what restaurateurs thought would happen, at the end of December). So, below, are outcomes:

Figure 1: New York City minimum wage, in 2018$ (left log scale, blue), restaurant employees (full service, limited service, respectively SMU36935617072251101SA, SMU36935617072259001SA), in 000’s, s.a. (right log scale, red). Deflation using CPI for NYC-Newark-Jersey City, n.s.a. Source: BLS, and author’s calculations.

If we focus on the 2019 increase — the topic of the Liberty Nation article — the prophesied calamitous drop in employment didn’t occur. In fact, employment stabilized. Of course, the data will be revised, and the collapse could occur in the coming months. But, for now, no cataclysm (much like none in San Francisco, nor in Seattle, …

More broadly, the (log) level of restaurant employment rises with the (log) real level of the minimum wage [AND SIGNIFICANTLY SO] although this is likely spurious correlation.The first difference shows absolutely no significant association.

No substantive changes in conclusions occur relative to the original post.

Bottom Line: Quote survey results from industry associations at your own risk.

Second Bottom Line: Recheck your data before posting. Thanks to commenter AS for highlighting the error, and apologies to all.

21 thoughts on “Self-Professed Policy Analyst Predicts Minimum-Wage Induced Employment Disaster [CORRECTED]

  1. Bruce Hall

    I didn’t comment in the spitball throwing contest section of the original post, but just two observations:
    1) – [obvious which you may just want to correct for the record] the minimum wage is left scale
    2) – As you said, may be too early early to say, but either there was a slowdown in the restaurant business late in 2017 or the bump from $11 to $13 may have been the “tipping point” for hiring/retention and the second bump from $13 to $15 simply continues the slide which looks to be about 2% or so cumulatively. That may be stabilizing or it may be a slow fall off until some minimum operating level is reached upon which growth might start again as long as the economy looks good.

    NYC should be as capable as any urban area to support a $15 minimum wage. To the extent that it causes a slowdown there, one might expect a $15 minimum wage to have a more profound effect in other cities with overall lower wages.

  2. pgl

    We should give AS credit by noting his entire comment:

    “Since you asked, the number of restaurant employees in full and limited service restaurants seems to be significantly overstated. If you look at labor statistics for the nyc region, specifically at the spreadsheet in “NYC Current Employment Statistics Latest Month,” you see that in Jan 19, full service restaurants had 161.3 thousand employees while limited service restaurants had 103.6 thousand employees, for a total of 264.9 thousand. Menzie has almost 600K employees, so something seems very wrong.”

    NYC boosts having a population of 8 million but then a lot of people from the burbs and New Jersey come here during the day to work. But no – 0.6 million people are not needed to feed us in eating establishments. So 265,000 sounds more reasonable.

    But note even with the right data – we are not seeing any significant drop off like the intellectual garbage peddled here by Princeton Steven. But even if there were a 10% decline, we would be talking about a mere 2650 jobs in a city of 8 million. We lost a lot more than that when Amazon wants not able to pull off that big tax dodge in Long Island City. And those were higher paying jobs. But NYC will be AOK!

    1. Bruce Hall

      pgl, just a correction to your comment: 10% decline would be 26,500 jobs (close to the Amazon promises). Of course, it’s only speculation that many jobs would be lost. It appears that about 5,000 jobs have already been lost since the minimum wage was increased to $13 per hour. However, as I stated earlier, “either there was a slowdown in the restaurant business late in 2017 or the bump from $11 to $13 may have been the “tipping point” for hiring/retention and the second bump from $13 to $15 simply continues the slide which looks to be about 2% or so cumulatively.”

      It really difficult to tell if this was caused by a business slowdown unless someone has data specifically related to the revenues/profits of those businesses employing the employees in question.

    2. Rick Stryker


      You are actually quoting my comment, not AS’s. AS had asked about the discrepancy, something I had noticed too, but I had just decided to drop this issue. However, since no one answered his question, I did, pointing out that Menzie’s numbers can’t be right if you cross reference with NY Dept of Labor data.

  3. pgl

    “Bottom Line: Quote survey results from industry associations at your own risk.”

    Of course Princeton Stephen told us over and over again that this was a “study”. If Princeton Stephen does not know the difference between an actual study and a self serving survey from a lobbyist group, he is dumber than we give him credit for.

  4. Moses Herzog

    This is why Menzie has my respect and admiration. Because he is willing to admit (the very rare ) error that ever arises from Menzie’s breakdowns. He “came clean” which is what good teachers and great policy researchers do, always double back and look for error. The thing to focus on here is, the end result as far as policy prescriptions, hasn’t changed.

    This is one of many reasons why I love this blog. Ego doesn’t get in the way of the work or the end goal with Menzie and Professor Hamilton. I am guessing Menzie is beating himself up a little over this—but he shouldn’t beat himself up. There is many a policy analyst or researcher who keeps plowing on after error (remember the gentleman Professor Hamilton busted on his use of logs?? still intransigent to the end). I know Menzie is a perfectionist, and he’s going to rip himself internally on this more than any of us would do.

  5. pgl

    “There is many a policy analyst or researcher who keeps plowing on after error”.

    Amen to that. Exhibit A: Princeton Stephen.

    1. Moses Herzog

      @ King John
      Thank you for sharing this link. There are other good ones of Krueger in youtube. I was looking for one of his last ones, which was on “universal income” given at Stanford University, but I cannot find it. Do you know where it is?? If you or any other commenters know where that Stanford lecture is in youtube or free links elsewhere I would be grateful. I don’t need it for studies or anything, just would love to watch it.

  6. Steven Kopits

    “…the prophesied calamitous drop in employment didn’t occur. In fact, employment stabilized” That’s how you interpret that graph?

    Here’s my interpretation:

    All is not well in the restaurant business. New York government’s decision to raise the minimum wage to $15 / hour by steps has eviscerated the full service restaurant sector.

    From the end of the Great Recession in 2009 until mid-2015, the full service restaurant sector added approximately 8,600 jobs per year. The increase of the minimum wage to $10.50 / hour in 2016, however, slowed growth to 4,300 per year.

    When the minimum range was raised to $13 / hour from the beginning of 2018, the full service sector cratered, with employment falling by 8,000 (-4.6%) from its November 2017 high through January of this year. By contrast, the full service restaurant sector lost only 2,700 jobs (-2.4%) during the Great Recession. Put another way, the effect of raising the minimum wage in 2018 was twice as bad as the harm incurred in the biggest economic downturn since the Great Depression.

    The cost is greater than just the jobs losses, however. Under normal circumstances, we would have expected the full service sector to grow by 6,000 – 8,600 jobs in 2018, which was the strongest year for the economy in quite some time. Consequently, these forgone jobs have to be added to the total. In all, an increased minimum wage can be credited with the loss of up to 16,000 jobs in the full service restaurant sector from late 2017 through January of this year. That’s two-thirds of the total employment Amazon promised to New York City over a number of years.

    1. PeakTrader

      A reduction in employment is expected from a substantial increase in the minimum wage over a short period.

      I’d expect poorly managed restaurants either to become more efficient, and therefore competitive, or go out of business, while more efficient restaurants find ways to become even more efficient and gain market share from the weaker restaurants.

      A very high proportion of restaurant workers will be much better off earning much more per hour, even with reduced hours. The positive income effect, particularly with workers with high marginal propensities to consume, will increase demand for products and generate jobs throughout the economy. Also, productivity is expected to increase, including lower turnover rate, and prices will rise only slightly. Moreover, I’d expect people out of the workforce, with higher reservation wages, to come back in. Furthermore, tax revenue will rise and government spending will fall.

      Overall, a subsistence wage will be a net benefit to the economy.

      1. Steven Kopits

        Disagree, mostly.

        A higher minimum wage will increase efficiency, absolutely. It will do so by focusing employment on higher value employees — experienced waiters over bus boys, for example — and by encouraging automation where possible. And it will shut down outlets of chains whose business models cannot accommodate higher wages. All true.

        It will do this by reducing activity in the sector and putting restaurants out of business which otherwise would have survived. You may consider government policy which closes down businesses artificially as a benefit. I do not.

        In general, customers’ budget constraints are unchanged with a change in minimum wage laws. So if you had $100 per week to spend on eating out on Dec. 31st, you probably still have $100 on January 1st. This will grow with inflation and the economy, but in the short term, it is largely fixed.

        Faced with higher meal prices, you will spend about the same, but eat out less often. That means fewer meals, and therefore a reduced need for cooks, waiters, etc.

        Now, who loses their jobs? Entry level workers, less skilled workers, and workers who have other issues, eg, drug use, a police record, failed to finish high school, poor work discipline, a physical disability or poor public presence, that is, first and foremost statistically that’s most likely to be young black men and the handicapped. Under this law, a young black man with a questionable record is worth either $30,000 / year, or zero. So if you’re a white girl from the suburbs, this is good for you, because you can squeeze out your young, black male competitors and prevent them from ever getting entry level jobs.

        For me, this is pure apartheid law. Sure, it’s politically popular, but in practice, there is probably no better means of discriminating against the weakest members of society than an unrealistically high minimum wage law.

        1. Steven Kopits

          To wit:

          “…we find that raising the minimum wage increases property crime arrests among those ages 16-to-24, with an estimated elasticity of 0.2. This result is strongest in counties with over 100,000 residents and persists when we use longitudinal data to isolate workers for whom minimum wages bind. Our estimates suggest that a $15 Federal minimum wage could generate criminal externality costs of nearly $2.4 billion.”

          This has intuitive appeal. The most vulnerable, as we note above, may be effectively deprived of a job in a high minimum wage environment, prompting some of them to turn to crime as an alternative form of income.

        2. PeakTrader

          Steven Kopits, you seem to believe businesses should exist without any labor standards.

          I’ve explained before, meal prices will rise much less than the higher minimum wage. It won’t cause people to go hungry.

          I’ve stated before, employment will fall, initially. However, it’s temporary, until labor is absorbed.

          With one-dimensional thinking, no wonder you believe it’s not a net benefit.

      2. pgl

        Someone missed the point entirely here. Card-Krueger’s 1994 paper? You never read it obviously. And all the econ bloggers are talking about Alan Krueger today. Poor Peaky – he has no clue with the conversation is even about!

    2. pgl

      Given your track record of intellectual garbage – NO ONE care about your “interpretation”. No Stephen – ain’t reading your stupid blog – EVER! BTW – see Menzie’s sector X post. You have been duly mocked!

  7. Grunschev

    Staffing levels have almost nothing to do with wages, everything to do with sales. If a business can do without that extra employee after the minimum wage rises, they could do without them before. Why would anybody hire people they don’t need to service demand? That’s just stupid and goes against your basic tenet of capitalism: maximize earnings. Companies don’t hire because taxes go down and they don’t fire because wages go up. They hire and fire based on sales. Period.

    It may be that a company will need to increase prices due to wage increases. And that may, in turn, reduce sales, which may mean staffing can decrease. But you certainly won’t see this the day the minimum wage goes up.

    This is business 101, folks. Not rocket science.

    (I’ve been building budgeting and forecasting systems for a long time. Never seen ANYONE base staffing levels on wages. EVER.)

Comments are closed.