Friday, May 20, 2016

Overtime

Like most economists, I was a bit baffled by the Administration's announcement of stricter overtime rules. The WSJ, and Jonathan Hartley and many others cover the obvious consequences on jobs, business formation and destruction, and so forth. A bit less mentioned, it reduces employee flexibility. If you like working more hours one week and less the next -- perhaps you have child or parent care responsibilities -- you're going to be stuck working an 8 hour day.  It's part of the general regulated ossification of American employment. Or, it could be one more inducement to substitute machines for people or make people independent contractors.

Why are they doing it? The government says it wants more jobs, yet there is no area in American life with larger impediments between a willing employer and employee than labor.

I'm trying to bend over backwards to understand a worldview under which this is a sensible idea.

One possibility. Suppose this is your image of work: Take as given that a person has a job, and the employer will keep that job going, and won't change the terms of the job -- lower the base wage, allow people to take overtime, etc. Take as given that the terms of the job are a pure bilateral negotiation, and there is money somewhere to absorb extra costs without raising prices.  Take as given also that the worker is in a bad negotiating position, and you, the benevolent central department of labor, care about moving this negotiation in the worker's way. Then, a rule like this is a way of strengthening the worker's bargaining position and driving some resources the worker's way out of the employer's pocket.

The counterargument is really just that all this "take as given" is false.

Here is an effort to put that debate in econ 101 supply and demand diagrams. Let's think of the rule as a mandated higher wage, like a minimum wage. The classic analysis says you get fewer jobs.


Now, how might you not lose jobs? The implicit assumption in my paragraph above is that the labor demand curve is vertical. Employers will hire the same number of people for the same hours no matter how much they have to pay. And they'll all stay in business too.

If that were the case, as you see, we wouldn't lose any jobs. There would be some unemployment, as more people want to work or employees want more hours than they can get. But I think advocates of these policies don't mind. Getting people to go out and look for jobs might not be so terrible.


Another way to apply econ 101 is to think of the new rules as new costs imposed on the employer. If employers have to bear more costs, their demand for workers drops down by the amount of the extra costs. Again, adding costs reduces employment. Once again, what are they thinking?


Well, again, suppose that the demand curve is vertical. Now employers simply bear the cost, grumble, but there is no reduction in the number of employees and hours.

Of course, with the assumptions made bare, we can think of lots of reasons that demand curves do slope, employers cut down on workers if they have to pay more direct or indirect costs, and companies don't have infinite funds coming from nowhere. But perhaps by showing implicit assumptions, there is some room for discussion that gets somewhere. I would be interested in hearing serious defenses of the vertical demand curve assumption.

Update: Good grief, Noah,  of course "to understand the true impact of overtime rules, we probably have to include more complicated stuff!" Who ever said otherwise? Isn't "econ 101" clear enough that this is a an extremely simple starting place? And aren't you the guy complaining about excess mathiness,  big black boxes in economics, and people who don't even try to understand the opposing arguments?

57 comments:

  1. Google "Fight for 15" if you really want to get mad, John... :-)

    ReplyDelete
  2. Why are you using Econ 101 rationale for an emotional argument from an emotionally dysregulated autocratic regime that has a long history of manipulating the populace (a.k.a. social engineering & vote buying)? Just come out and say it, the minimum wage and overtime arguments are not rational arguments, they are emotional appeals made by elitists who talk down to the electorate, knowing that we're at the breaking point where more people depend on the government than actually fund the government.
    It's democracy in action, like a sheep and two wolves voting on the dinner menu.

    ReplyDelete
  3. This only applies to people at the low end of the pay scale, keep that in mind. The exploit that service industry employers have been using is to declare a $12 an hour employee a "manager" and ask them to work 60 hours a week without overtime, opening or closing the store, or keeping it open for cheap during low business hours. I know people being taken advantage of this way. These people don't want to work 60 hours, but they are being forced to do it to keep their job. This reclassifies low paid supervisory personnel at retail and fast food places are non-exempt for overtime.

    ReplyDelete
    Replies
    1. Very good explanation Mary. I do expect the ramblings and "econ101" nonsense to continue here though.

      Delete
    2. Right. The rationale for not giving managers overtime is because they are assumed to be getting good compensation and other perks, be it frequent-flier miles from business travel, stock options, a path into the executive ranks, a high degree of autonomy over how they handle their work.

      The extra, often unscheduled time they were putting in was compensated in other ways.

      Applying this rationale to a glass-ceilinged fast-food worker making 1.6x minimum wage was never the intention. They are directed to do "whatever it takes" to get the job done and can't plan their days or weeks. Yet unlike the traditional white-collar manager, they get no perks or status to make up for this.

      Delete
    3. Why would this change anything? If they were paying some guy $37K a year to work 60 hours before the law, why wouldn't they lower his hourly pay so that they still get their 60 hours and he still gets his $37K? If he could find a better deal, why wouldn't he have already taken it?

      Delete
    4. To Mark Barbieri,

      Yes. The argument for government control is that a free market and competition for workers is not assigning the "correct" wage or salary to people.

      Starting from that assumption, then the government should set all wages and prices. Throwing in this or that employment rule is an unending process which will in time have that result, detailed control of wages and prices.

      That has worked out so well (sark) for Russia, China, Cuba, and Venezuela. I wonder what is stopping the US from doing the logical and correct thing. That is, control everything and achieeve utopia.

      Delete
    5. Being in the restaurant industry, I'll tell you what fast food is going to do. They're going to go to one salaried manager instead of two. They'll pay the premium for that manager. They'll eliminate the other job and spread it among two 'key employees' that make $10/hour instead of $15 and aren't salaried. Those two will never get more than 40 hours. In addition, you're about to see automated burger makers and order takers to further lower labor expense to keep total ROI constant.

      In casual dining, the assistant, salaried manager is probably also going the way of the dodo with this regulation change. We'll do the same as the fast food guys, though we'll pay slightly more (maybe even the same hourly wage as before), but we'll try to get them down to 31 hours a week so that we don't pay them benefits. Thus, we'll increase total hours, decrease the take home pay of 2/3 of our management and cut the benefits of 2/3 as well. Big win for 'the people'.

      Delete
    6. Mary, you're talking to economists. Why on Earth would you cite the actual empirical facts driving this policy in the real world? Do you not understand what economics is? Did you not see the pretty graphs above that show that the writer's model makes predictions that are consistent with his model?

      Delete
    7. Nothing that Mary said was an argument against what the Professor stated in the post. Econ 101 teaches that there will be adverse reactions to this law unless there is a vertical supply or demand curve. Wishing that the demand curve was vertical does not make it vertical no more than wishing gravity did not exist. Mary's argument is equal to someone saying man made CO2 does not contribute to climate change because I need to drive my car.

      Delete
  4. I don't think of this as a way to increase wages. I think of it like a law mandating stores be closed on Sunday: obviously the stores earn less, but without it, there's no way you can stay closed on Sunday, because of competition. You're moving from an equilibrium that favors strivers to one that favors slackers. As a slacker, the only way I was able to make the current system work for me was by saving up the pay from the mandatory overtime until I could quit working altogether. I'd rather work 40 steadily than 60-then-0, so whatever hurts the current mandatory-60 system helps me.

    ReplyDelete
  5. Here is one off the wall proposal.

    There are substantial training and hiring costs for new workers and experienced workers are often more productive than new workers. Also there is often a coordination cost to be paid to get multiple workers to jointly realize certain tasks.

    As a result there are many situations it is more appealing for employers and willing existing employees to pay those employees more to work more hours than it is to hire new workers.

    However, the utility received per dollar is logarithmic so in terms of utility society is much better off employing more workers even if this comes at a net cost to overall efficiency and wages paid.

    ReplyDelete
    Replies
    1. Peter: Your logic was the same used by countless Indian governments to justify using manual labor to build dams and highways. This is because job specific expertise and training are just a form of investment by employers that make employees more productive. And the transactions costs of hiring, training and managing the next employee are significant.

      Delete
  6. I don't know how plausible this is, but you could make an externality argument. Sometimes, workers work overtime in order to signal to their employers that they are good workers. But by engaging in this signaling game, they are increasing the cost of not working over time. Effectively, over time laws are attempts to staunch the "rat rate" of workers trying to out-compete each other in time worked, despite diminishing rates of productivity.

    ReplyDelete
    Replies
    1. You could make that argument. But what are we doing, coming up with new theories after the fact to justify policies? Shouldn't we come up with clever new theories, verify them in the data, and then start thinking about policies? Way too much economics consists of coming up with clever theories ex post for policies that have been pursued for other reasons.

      Delete
    2. To John Cochrane,

      In complete support. The US has been reduced to Rulers directing the Peasants. This is evident from how little explanation (none) that we peasants get from our rulers.

      It is ludicrous to guess what is behind government policy. Our governemnt does not bother itself with issuing written explanations and justifications. Verbal explanation is limited to "These policies will strengthen the common man".

      The big task is complete. The public now accepts no explanation for policy. Now, we can be ruled by edict, with public response reduced to "The government must have a good reason".

      It is futile to "figure out" policy. At best, this presents an opinion. Supporters of government policy may always say that any criticism does not necessarily mean that the policy is not good overall. It would be better for all thinking people to complain loudly that it is an outrage for the government to act without detailed explanation and justification.

      Cry out for the government to publish its analysis. Then, that can be torn apart as the idiocy that it is, almost always.

      Delete
    3. Why are tournaments prisoner's dilemmas? They might be if you only consider the workers, but if you consider the owner and and customer, who benefit from having the best employees advance, I doubt it is.

      Delete
  7. Your arguments are based on partial equilibrium models. Most people who believe in the minimum wage and so forth have some sort of Keynesian type model in their head. The underpinning of this model is that if you redistribute income to the poor, then demand curves shift outward and employment will increase. I do not believe in this model, but you will need to confront their arguments at this level and not rely on a partial equilibrium analysis

    ReplyDelete
    Replies
    1. There is substantial evidence that low-income workers/families have higher propensities to consume than do high-income workers/families. In other words, increasing income to the poor (by redistribution or by increased hiring of 'poor' workers) will lead to higher demand for consumption goods and services.

      Whether you 'believe' it or not does not change this relationship.

      Delete
  8. Is the demand curve functional related to the productivity level? If so ...and there has been a significant divergence between the productivity level and the wage level over an extended period of time - say 40 years or so ... then would the employer demand curve be vertical until the productivity level and the wage level equalize? At that point the demand curve would become "normal" = downward sloping to the right...???

    ReplyDelete
  9. Yes, the right-wing roars against minimum wage and OT rules...but mute when it comes to property zoning and criminalization of push-cart vending.



    ReplyDelete
    Replies
    1. I don't know who you are calling "right wing" or "mute," but I and just about every other free-market economist and institution is loudly against zoning, restrictions on push carts, ubers, occupational licensing, and so forth.

      Delete
  10. Here is my full name David Vernal Wold...retired ... doctor of arts in poly sci with sociology and economics as subareas ....total 112 grad credits ...taught soc for about 15 years ... all done at Idaho State University

    ReplyDelete
  11. .
    This article suggests to me that its author makes more than 50K.
    .

    ReplyDelete
  12. When I took economics back in the 1970s we were taught supply and demand curves and told that the only place where one might find such in an actual real world market was the one involving the sale of marijuana. All other markets were heavily influenced by power relationships, regulations, structural issues and so on. This was MIT. There was no point in lying to engineers.

    It might help one's understanding if one simply looked at the problem from the point of view of an employer rather than invoking a long discredited approach to understanding markets. It has been 40 years. My sliderule is in the store room. I use a calculator or even a computer nowadays.

    If you look at the NHANES time usage studies since the 1950s, one obvious pattern leaps out, aside from people watching television instead of spending time with their friends. Exempt employees have been working longer hours while hourly employees have been working shorter hours. If something costs money, people will use less of it, while if something is free, people will use more of it. Over the years, inflation has increased the number of free hours employers can demand from their employees. If you are looking for a horizontal or vertical line, that's one place you will find it.

    There are a number of things an employer can do in response to the exempt/hourly line being raised to compensate for inflation:

    (1) Employers who already have hourly employees who are working 20-30 hours a week could cut the hours of their formerly exempt employees to 40 from 50-60 and give some hourly employees an extra 5-10 hours a week. This puts more money into payrolls and boosts demand.

    (2) Employers could simply pay the overtime. If an employee is working 60 hours for $X, their pay would go to 1.25 * $X. Again, this would increase demand. Alternatively, they might raise wages to take those employees back over the new magic line at which marginal hours become free.

    (3) Employers could rethink their business to use less labor. This is called either good management or total factor productivity. As in the old USSR, free goods tend to be abused, so putting a market price on labor would improve productivity.

    (4) Employers could invest in equipment so they use less labor. This is supposedly at the heart of capitalism, but when labor is free, investment has minimal return.

    The problem is much like that in Lenin's Soviet Union. When a commodity, bread or labor, is free, it is likely to be used inefficiently. Putting a price on bread or marginal labor hours will inconvenience some. It will no longer be economic to feed bread to one's pigs or to have employees working 50-60 hour weeks.

    There is a lot of concern with the low rate of productivity growth in our economy. Raising the exempt/hourly line would encourage more productive use of labor. There is also concern with stagnant wages over the last forty years. Again, raising the exempt/hourly line would encourage higher wages in the form of longer hours and overtime pay. There is a trade off between raising productivity and having a bigger payroll, but this measure seems to be putting pressure in the right direction.

    ReplyDelete
    Replies
    1. Is this why we're seeing less productivity increases?
      If labor is free why try to find a better way to do the work?

      "4) Employers could invest in equipment so they use less labor. This is supposedly at the heart of capitalism, but when labor is free, investment has minimal return."

      Delete
    2. You make it sound as if there is no extra cost to asking an exempt worker increase their hours. If that is the case, why aren't they all working 24/7?

      Delete
    3. Mark, this argument is about a local, marginal cost. Sure, there probably is eventually a substantial cost to asking an exempt worker to marginally increase their hours, because as you get closer to 24/7, presumably the worker would quit and find a different job. But that is different than the marginal decision to work an extra hour "occasionally" when the worker is at 40 or 45 hours. The worker might not be happy about it, but is unlikely to quit - it's smaller than the cost of changing jobs. So the employer does get the marginal hours "for free," as long as the demand is within what seems "reasonable."

      Context: I'm a salaried worker paid less than the new threshold who's been told to report 8 hours a day despite regularly working about 9 hours and intermittent instances of being explicitly asked to stay 10 - and who has done so, despite knowing that I'm not getting paid for it. Why? I want my employer to think I'm dedicated. I'm not going to get a big bonus (we're not for profit). But I want to build a good reputation.

      All that said, I don't love the new rule. At the very least, why not phase in the increase as is often done with minimum wage increases?

      Delete
  13. I think you are missing a key point: Actually many economists who favor the minimum wage and other rules such as the overtime rule, don't disagree with the "economics 101". They just recognize that it is not the sign of the demand curve that matters---it is also the slope. Evidence says that the slope is sufficiently steep at current wage levels that the benefits flowing to lower wage workers far outstrips to costs of the unemployment induced.

    ReplyDelete
    Replies
    1. I've never seen that argument made by an economist or anyone else. Please show me someone who advocates a higher min wage that admits it will lead to job or hours worked losses.

      Delete
    2. Here (http://www.alternet.org/economy/robert-reich-7-reasons-why-minimum-wage-should-be-raised-15-hour) Robert Reich says "a $15/hr minimum [wage] won't result in major job losses" implying that some job loss would be an acceptable cost of raising the minimum wage.

      Delete
  14. Well, I do agree that non-exempt workers tend to work full time hours carefully scheduled to avoid triggering overtime costs; while exempt workers tend to work more than a 40 hour workweek. So, seems clear that this will even out hours worked between the two groups more than in the past.

    Two other alternatives besides those identified by Mr./Ms. Kaleburg:
    - Greater differentiation in wages - much, much more of the total rewards will be incentive compensation, based on performance and results, or
    - Lowering benefits spend.

    In the first, you announce next year's direct compensation increases effective today, but, you confirm that increased compensation will be in the form of incentive compensation, that base salaries must be lowered to offset the impact of greater overtime spend, then allocating the increased total spend based on performance/results. You agree to confirm the amount spend, for all workers as a group, before and after the change. You confirm to workers that you will pay MORE than you currently pay, for all workers subject to overtime taken as a group, but that the allocation must meet new government requirements for overtime, etc. As a result, each and every employee may earn more or less than they currently earn - in order for the employer to comply with the new rules.

    In the second, you recognize that the benefits spend is already very heavily weighted in favor of lower compensated employees who are subject to overtime. Where a health plan charges each employee the same dollar amount for single coverage, the employer spend is a much greater percentage of total rewards for the lower paid. So, raise employee contributions as necessary to offset the greater overtime costs - but, for each employee, not beyond the 9.66% of pay necessary to pass health reform's "affordability" requirements.

    After two years, you end up with about the same "total rewards" expense - but allocated in a much more close relationship to results and performance. You also confirm to employees that your preference was the total rewards structure that used to exist - before the health reform and overtime mandates, confirm that you are only complying with the law, and give each and every employee the name and phone number and email addresses of their elected representatives, the white house and the department of labor, so they can call in to voice their support for the new allocation rules.

    ReplyDelete
  15. I am a low wage worker who married and moved to Canada where I can now work 44 hours before I hit overtime, which means I can work 44 hours now because my employers don't pay overtime. Life is so much better. I have a kid and family to feed, let me work!

    It's just plain stupid to have someone work overtime when you can just grab another unskilled worker to work the extra hours. As a result, there were many times when I would have loved to work a 50 or 60 hour week, but law make it financially insane for an employer to work me that many hours. The only way for me to work the number of hours I want to work is to get a second job.

    ReplyDelete
    Replies
    1. Although you say that it is "plain stupid", there are downfalls when hiring unskilled workers. Looking at the costs to pay for multiple unskilled workers, and the costs to pay for the current workers, would exceed the amount the business would spend just to accommodate for both the new, and the currently employed.

      At least with overtime, schedules would be more flexible as stated by John. Not only that, but if businesses would hire an unskilled worker, the time it would take to teach the worker how to perform his/her job would be unpredictable; and may be money that is lost if the worker is incapable of handling the job.

      Delete
  16. 10 hour days are better for me too. Let me get my work done so I can have an extra whole day off to actually do something.

    ReplyDelete
  17. To the point of Econ 101 rationale as well as the emotional part, I would argue you should look at what each party wants to get at least a first look at the disparities. I did a first attempt using Econ 101. Feel free to roast. https://app.box.com/s/y3fl9tccy6vto861hyoar9zjy190djnm (it's a GIF picture of Supply and Demand curves. Be afraid, be very afraid).

    ReplyDelete
    Replies
    1. The point is that arguments such as "minimum wage" and "mandatory overtime" are not rational arguments worthy of economic analysis because from their very inception, they are emotional pleadings akin to the matador's cape. They are designed to distract and appease but offer no real solution to anything. Where the free market functions efficiently, the central planning of governments with arbitrary and capricious dictates interferes with normal signals and incentives. These are just pandering promises to solicit votes and grant even more fiat power to elite power organs. Giving them full economic analysis, as elegant and beautiful as the attendant logic may be, gives these wedge issues credibility as actual logical arguments.

      Delete
  18. I think a lot of this is over analysis. This is, I think, the most straight forward example of employers cutting back other compensation. Since this is basically switching employees to hourly from salary, many hourly (most?) do not have benefits like sick time and vacation. Another way around this, though I do not know if it is addressed in the law, why is it assumed employers will divide a salary by 40 hours * 52 weeks? If I were an employer I would calculate the average overtime for the past year, multiply by 1.5, add to 40, then multiply by 52. The second way would result in no benefit to the employee and no additional cost to the employer.

    ReplyDelete
  19. Overtime helps match opportunity costs for workers (which you ignore in this post) with wages. As time worked per week increases, so do opportunity costs. Unpaid overtime means that as time worked per week increases, average wage decreases.

    I'm not an economist, but it only took me a few minutes to think of this. And it took only a few minutes more to find the same idea in a review of The Economics of Overtime Working by Robert A. Hart.

    "First, governments may attempt to reduce the health and social costs of excessive work demands by imposing high labor costs on firms’ marginal daily or weekly hours beyond acceptable norms. Second, labor unions may require high marginal rates as compensation for potential adverse effects on their members’ welfare."

    The myopic focus on employers and employment ignores the opportunity costs to workers.

    ReplyDelete
    Replies
    1. Mike, its good to see you're still keeping these freedom loving libertarians on their toes ; ) However, no one is ignoring costs. The employer has to compete as well as the employee. The terms of the working arrangement has a total cost to both parties. Mandating additional costs to the employer will have the effects John is citing.

      Delete
    2. First, Cochran is ignoring the opportunity costs to workers of longer hours. If you disagree, then cite him please.

      Second, "the effects John is citing" are based on simplistic economics 101 analysis, which has patently failed to accurately model the results of increases to minimum wage: why should we believe in it here? Where is the EVIDENCE?

      Delete
  20. I am not saying this argument is correct, but one argument one could make about how increasing overtime increases jobs goes like this: An employer has 8 employees working 50 hours a week; i.e. 400 hours of work. Overtime goes up, so employer hires two more employees to avoid paying the overtime. Voila, 2 more jobs.

    ReplyDelete
    Replies
    1. That was the exact impetus behind the overtime laws when instituted in 1938.

      "Proponents of the bill stressed the need to fulfill the President's promise to correct conditions under which "one-third of the population" were "ill-nourished, ill-clad, and ill- housed." They pointed out that, in industries which produced products for interstate commerce, the bill would end oppressive child labor and "unnecessarily long hours which wear out part of the working population while they keep the rest from having work to do." Shortening hours, they argued, would "create new jobs...for millions of our unskilled unemployed," and minimum wages would "underpin the whole wage structure...at a point from which collective bargaining could take over."
      https://www.dol.gov/oasam/programs/history/flsa1938.htm

      Delete
    2. An 8 employees taking a pay cut. If their productivity dropped by 20%, why would the employer not adjust their pay accordingly?

      Delete
  21. I'm pretty sure some derivative of this argument was fought when the 40 hr work week was instituted, or the implementation of a 2 day weekend. Things that in hindsight were very valuable improvements to our labor economy.

    ReplyDelete
    Replies
    1. Why do you think those were improvements? They hurt workers who want to have Fridays off instead of Saturday or who want to work 44 hours per week.

      Delete
  22. The White House fact sheet declares that the new rules will increase wages for workers by $12 billion over 10 years or $1.2 per year. As far as I can tell, direct wages paid to workers was $7.1 trillion in 2013; if I adjust for GDP increases since then, I get $7.8 trillion now. So the WH is telling me that the new rules will increase wages by .015%. How in the world did they measure this at this level of accuracy? I don't think they took into account compliance costs, which reduce the monies employers have available to pay employees. Given the wide scope of this rule, I wouldn't be surprised if compliance runs into the billions, if not the tens of billions, wiping out any salary increases and making a major PITA for all concerned.

    ReplyDelete
  23. John,

    "I'm trying to bend over backwards to understand a worldview under which this is a sensible idea."

    Perhaps you can explain this:

    https://research.stlouisfed.org/fred2/series/AWHNONAG

    Average weekly hours of production and non-supervisor employees (all private industries) - 33.6 hours.

    https://research.stlouisfed.org/fred2/series/AWHMAN

    Average weekly hours of production and non-supervisor employees (manufacturing) - 41.9 hours.

    Pick your labor supply / demand curve and then explain why there is a large discrepancy between hours worked by employees in the manufacturing sector and hours worked in other sectors.

    I am not saying that the Obama administration's approach is a sensible idea. But this should address what the Obama administration is trying to achieve - narrowing the gap between hours worked in one sector versus another by introducing a significant cost wedge incurred by all.

    ReplyDelete
    Replies
    1. Frank,

      Could you clarify the argument that you are making here? I'm not quite sure I understand how this rule would narrow the gap between hours worked in one sector and another.

      According to the White House Fact Sheet on this new rule, this rule applies to full-time salaried employees and is intended to counter the drop in share of covered full-time, salaried workers from 62% in 1975 to 7% in 1916.

      But looking at your numbers, I'm not sure I can see where the effect comes in. https://research.stlouisfed.org/fred2/graph/?g=4xct (consolidated the time series, added earnings)

      First, manufacturing employees are more likely to be unionized. Those labor agreements tend to restrict the hiring of new labor and come with their own overtime rules. So this rule wouldn't have much of an effect there; that conclusion is, I think, supported by the manufacturing time series you linked, which shows no significant change in the average weekly hours in manufacturing over the last 50 years. (If this is what you were implying by comparing the two time series, then I concur with this part.)

      Second, it's not immediately clear to me how the the drop in average weekly hours is due to the erosion of this rule's benchmark or how this rule would reverse it. Google suggests that the last post-1970s change was in 2004, but that Total Private line does not have a break there. In fact, it has only changed about an hour since ~1992, even though the average hourly earnings had been rising pretty consistently for the almost-20 years before.

      Thirdly, if the 55-percentage-point drop in coverage rate was the cause of the decrease in the average weekly hours worked, I kind of would have expected the drop to be more than ~2 hrs (since 1975).

      These data are highly aggregated (e.g. the rule covers salaried employees but the time series include willingly part-time employees), but I would still expect to have seen an effect there if the rule had had one.

      (I leave it to the blogger to find the part-time fraction of employees by industry through time and to compare that to the average weekly hours worked to determine whether the effect is solely in demographics. I have no suggestion on how to determine "nonvoluntary" part-time work.)

      What am I missing?

      Delete
    2. Anonymous,

      "According to the White House Fact Sheet on this new rule, this rule applies to full-time salaried employees and is intended to counter the drop in share of covered full-time, salaried workers from 62% in 1975 to 7% in 1916."

      I presume you to mean 62% in 1975 to 7% in 2016 (this year).

      "First, manufacturing employees are more likely to be unionized. Those labor agreements tend to restrict the hiring of new labor and come with their own overtime rules."

      And I would imagine that the new federal regulations are more generous to workers than current labor union agreements?

      https://www.dol.gov/featured/overtime/

      The final rule will:

      1. Raise the salary threshold indicating eligibility from $455/week to $913 ($47,476 per year), ensuring protections to 4.2 million workers

      2. Automatically update the salary threshold every three years, based on wage growth over time, increasing predictability.

      In response to the new overtime rule, employers can:

      a) Pay time and a half for overtime work
      b) Raise worker salaries above the new threshold
      c) Limit workers hours to 40 hours per week

      Looking at the weekly earnings threshold, before it was about half the weekly average for employees to get time and a half for overtime, now it is close to the average.

      "Second, it's not immediately clear to me how the the drop in average weekly hours is due to the erosion of this rule's benchmark or how this rule would reverse it."

      The intent of the rule is not to reverse the slide in average weekly hours for all private industries. The intended effect (as I understand it) of this change is to reduce the weekly hours worked by manufacturing employees in the hopes that this translates into more jobs - more people working fewer hours. Like I said, I don't agree with the approach.




      Delete
    3. Anonymous,

      The big problem that I have is this statement by John:

      "I'm trying to bend over backwards to understand a worldview under which this is a sensible idea."

      John proceeds to then incorrectly draw his labor supply demand curves - with no units for either Wages on the Y axis or Work on the X axis and with no kink depicting the increase in wages (time and a half) for weekly hours over 40 worked.

      And the bigger problem (the one that I alluded to) is that a single labor supply demand curve (pick one from the four John offers) assume that all labor is ubiquitous, when clearly from the data presented in the links that I gave - all labor markets are not identical.

      Delete
  24. If profits are coming from economic rents, wouldn't the labor demand curve be vertical?

    I would assume the government thinks there's a fair argument to be made that there's quite a bit of room like this to maneuver. Profits are at record highs,as well as the shares going to capital vs labor, and wages haven't kept up with productivity gains.


    The other (more plausible, i think) argument- does it have to be vertical, or just sloped heavily enough? The argument for things like free trade/minimum wage usually go along similar lines. Even if you have to completely subsidize the worker who lost their job, and lose some money to inefficiencies, you can end up ahead.
    I think most people would be willing to concede that there are likely to be casualties.

    I don't think I've provided a very good argument why they're correct, but maybe at least shown it's not obviously trivially destructive.

    I'd be curious to see any data on what happened after the 40 hour workweek, or overtime was implemented, historically. (or better yet,any developing countries recently)

    ReplyDelete
  25. Can you respond to Noah Smith's rebuttal? He points out that you used the same model two different ways: 1) overtime as a wage floor and 2) overtime as a negative demand labor shock. He then states that the former drives wages up while the latter drives them down.

    Looking from the outside his argument seems compelling.

    ReplyDelete
  26. Re: "Isn't "econ 101" clear enough that this is a an extremely simple starting place?"

    It's an invalidly simple starting place. That's Noah's whole point. Mathiness isn't only about too much math, it's about an incorrect use of math to make some rhetorical point. In this case, "extremely simple" is most definite NOT a feature but an egregious bug. Think man.

    ReplyDelete
  27. Noah Smith points out that you haven't even used Econ 101 right.
    Also, jobs are not apples on a cart: the barriers to changing jobs are a large cost disparity between employees and businesses - for a business a change of one employee only alters the functionality of the business slightly, while for the employee the result is costly and fraught with peril. America's bizarre system for financing health insurance and pensions contributes to the need for further jiggering by the government to avoid further market inefficiencies and exploitation of disparities in management and individual employee power. The US is fairly unique in having a large fraction of net employee compensation split off from wages. The result is a different effective pay rate at the same nominal wages for salaried and hourly workers and an large stepwise cost for adding another employee. The fact that companies are able to shift costs onto the government by splitting employees into short hour workers and uncompensated overtime workers, and in the case of the Walmarts of the US pay less than food stamp wages is a another factor making Cochranes badly applied Econ101 model nothing but deliberately misleading.

    ReplyDelete
  28. John,

    "Update: Good grief, Noah, of course to understand the true impact of overtime rules, we probably have to include more complicated stuff! Who ever said otherwise?"

    You did.

    "I'm trying to bend over backwards to understand a worldview under which this is a sensible idea."

    You really "bent over backwards"? Would it have been "extreme" mathiness to includes units of measure for your X axis and Y axis? This is demanded in every other science that utilizes graphs to depict ideas - why not economics?

    Suppose you had chosen the units of $ / hour for wages and hours per week for work. What your graph should have depicted is a kink in your labor supply curve once a laborer is working more than 40 hours per week (time and a half).

    The answer isn't that the labor demand curve slopes up. The answer is that the labor demand curve becomes more positively sloped after the 40 hours per week threshold is reached (hence the kink that I am referring to).

    ReplyDelete

Comments are welcome. Keep it short, polite, and on topic.

Thanks to a few abusers I am now moderating comments. I welcome thoughtful disagreement. I will block comments with insulting or abusive language. I'm also blocking totally inane comments. Try to make some sense. I am much more likely to allow critical comments if you have the honesty and courage to use your real name.