The overtime boom

The last time U.S. factory workers put in longer weeks than they averaged in February, Rosie the Riveter was on the assembly line and American GIs were fighting Nazis in Europe.

All those extra hours helped to drive five straight months of manufacturing growth in the U.S., racking up 52,000 new factory jobs, according to Labor Department data. That includes 14,000 positions in February alone.

Good news of course, but there is a dark lining to the cloud.  I take the heavy reliance on overtime to be another sign of labor market polarization, and of low employer demand for a large number of the unemployed.

The story is here, hat tip goes to this chain.

Addendum: As Matt Yglesias reports, retail sales are strongly up too.


2 hours a week is very little overtime

If an employer is Keynesian average, do they hire or increase hours of existing workers (in the face of uncertainty)?

IOW, this statement "another sign of labor market polarization, and of low employer demand for a large number of the unemployed" is more likely than not bias.

In addition, hiring more workers means hiring more middle managers and what firm now permits such? They have now all been shown the door.

Who doesn't have a freeze on middle management hiring, today?

Software, computing sectors. Probably a lot of other tech sectors.

"Who doesn’t have a freeze on middle management hiring, today?"
Government, education.

Yes, that comports well with the massive drop in public sector workers

At my factory, an average work week of 42 hours does not mean each worker is working two hours of overtime. It means that line operators start working 5-6 twelve-hour shifts per week instead of four, while warehouse stays constant and material handling adds twelve hours every other week. So for me, the overtime increase is huge.

So why haven't more workers been hired?

Because the marginal cost of health insurance for overtime hours is zero, while the marginal cost of health insurance for hiring another worker is large.

It's true, overtime is a steal for the employer now that so much is in benefits. But also, last year we had a ton of 36-hour weeks where we shut down on the weekends, so it's hard to match staff to demand. Still, I think when they install a new production line, it would behoove them to also begin training operators to run it... especially since 2 people quit for every 3 we train, and they only train a group of three every 2-3 months.

I'm sure by now most employers have read all 3,000 pages of PPACA and have no concerns about its effect on hiring.

One need not drag the ACA into this. We should note that job creation has been occurring in low wage low skill jobs*. it's the middle tier jobs where it lags. But those are the jobs that generally do provide health insurance already, not because employers are charitable, but because the labor market at that level demands it. You'd have a better point if you directed it to the high cost of American healthcare, and indeed, this trend toward more overtime and more temps is nothing new; it's been in place for many years, and I agree that the fixed cost of health benefits is a big part of it.

* The assumption by those employers may be that their low-wage employees will qualify for Medicaid so they will not have to worry about the ACA's mandate. Or perhaps they regard the 2K/employee for not providing insurance as a tolerable cost and it does not deter them from hiring (as opposed to the much higher cost of paying even partially for health insurance which middle tier and high end jobs require through market forces)

So, your argument is that most employers have, in fact, read all 3,000 pages of PPACA and have no concerns about its effect on hiring? Glad we agree :)

Insurance is not offered at higher incomes because the labor market demands it, it is offered because of the tax incentives. Employers derive no benefit from offering it, employees derive no benefit from receiving it, indeed both are made worse off, as everyone gets stuck in a system with few incentives for cost control.

Lower income jobs don't get health insurance simply because the cost is large when compared to their monetary compensation.

My argument is that the ACA is superfluous: the cost of health insurance alone, ACA or no ACA, suffices to explain this trend (which is of long standing even if our host is just noticing it).
And if you don't think most middle class (and above ) workers regard health insurance as a non-negotiable and necessary part of compensation then I have to wonder what color the sky is in your universe. As for the tax break, it's not trivial, but it applies only to the employee: the employer is not taxed on employee compensation in any form; it's all the same whether he is paying out 10K in wages or 10K in benefits. But people go for health insurance even when they are taxed on it (yes, that happens: see domestic partner coverage). And more importantly, huge numbers of people, perhaps a majority, are not even aware there is a tax break (most people are not political junkies after all). I've found a surprising amount of ignorance when I've discussed this with people, in regards to the afore-mentioned case of domestic partner benefits which are fully taxed. In fact only my ex-boss and a lawyer friend even knew that health insurance is not usually taxed.
As for the employer benefit, yes there is one, or they would have canceled their employee health plan decades ago. They need to offer those benefits to attract the best workers possible-- surely that's not hard to grasp, as it's a straight out market argument. Skilled people expect health benefits and if one employer does not provide them, they will pass on working there and go with another employer. And if someone were to cancel the company health plan, they would find people quitting for other jobs, the best people often being the first to go.

Obamacare is a fixed cost. It's natural to try to avoid our by scaling the existing workforce.


But the growth in the labor markets seem to be accelerating as we approach the implementation of Obamacare. Certainly, there is little evidence of reduced aggregage hiring.

You need to look a little deeper. Consider the breakdown between full-time and part-time job number increases. IIRC, the most recent numbers were up, but the big increase was in part-time jobs. As I understand it, employees working less than 30 hours a week (ie. part-time employees) are not subject to the ACA.

So do benefits not only avoid taxes but also the overtime laws?

Would be interested to see this broken down by number of employees per firm, i.e. is this being driven to any extent by firms trying to avoid going over some statutory threshold such as 15 or 50 employees?

Retail sales declined from January without seasonal adjustment. They did not decline from Jan-Feb in 2012 or 2011.

Are you saying you think there's a new, recent seasonal trend that the published numbers are missing, or do you disagree with seasonal adjustments in general? Because the latter is just wrong, and the former seems unsubstantiated.

If you want solid information, read calculated risk - www.

Solid information is that unadjusted retail sales fell January to February, something that has not happened in the last three years.

So, put you down in seasonal adjustment is nonsense camp, then.

For those interested in actual numbers, graphs, and comparisons, then enjoy an introduction to how calculated risk works -

I expected the good headline number, but the report makes me thing the revised numbers will be lower.

The gloss on retail sales I read said that the increase was largely due to increase in gas prices, which doesn't strike me as particularly good news.

Retail sales less Autos was up 1%. Retail sales ex Autos and Gas was up 0.4%. Retail sales control group was up 0.4%.

From the employers perspective, more employees = more healthcare costs. More overtime = no (or very minimal) increased healthcare costs. I believe this is the calculation driving the increased overtime.

Todd wins the thread.

But historically, most employers provided healthcare to full-time employees. Is your point that because healthcare in general constitutes a higher proportion of wages now than it did in, say, the 1960s, that this is the main driver? Or are you just saying this trend is pretty normal and should be observable in historical post-recession data? (Both, I think, are valid.)

Health insurance costs $x/month, whether your employee works 120 hours or 220 hours in a week.

I was under the impression that high overtime rates were taken to be a leading indicator of higher rates of hiring. That is, companies were looking to expand production as much as possible before finally investing new capital in expansion. Why is it different this time? The Affordable Care Act doesn't seem like a reasonable scapegoat, since to be earning overtime, employees must already be full time. Historically, most employers provided healthcare coverage to full-time employees, so it doesn't seem like that would be impacting the situation.

Personally, I see this as more evidence that the Government's inability to set a long (or even medium) term trajectory for the country's finances is keeping companies from investing. The economy is moving forward, so more production is necessary, but it's at this point, it's easier to invest in overtime than new hiring/infrastructure. Maybe if a reasonable compromise with a long-term debt ceiling extension gets done this time we'll see companies push some of that capital reserve they've been holding back into growing their business.

I think the issue is that employers may not want to add more employees because working the existing ones harder doesn't increase the number of covered employees in the health plan. If they're already full-time, you'd want to work them as hard as possible before adding more people. It could be a leading indicator since these same companies may eventually need more people anyway, but you can theoretically increase production by 25% by having people work 50 instead of 40 hours and those extra 10 hours--which only impact wages and not benefits--may be cheaper than hiring more workers who get new fixed cost benefits.

Right, but this is true pre-Obamacare as well. Maybe not in the retail/fast food industries, but in manufacturing certainly.

Some commentors upthread are acting like Obamacare is the first fixed cost in the history of employment.

'of low employer demand for a large number of the unemployed'

Or, 'of low employer capability to train an adequate number of workers, leading to a large number of unemployed'

Why is this a surprise? If you believe in ZMP, then companies would be slow to hire workers. In terms of sticky wages, companies are sticky on hiring workers. In this productivity driven economy, you can't hire somebody and give them a shovel.

These OT hours will not convert to full time employees until employers are sure that demand will stay higher, and until they figure out the regulatory implications/costs of hiring additional workers.

My impression was that hours worked is a leads employment. It's a lot easier to cut hours or pay overtime than hiring or firing someone, and it's much easier to reverse if it is only a few bad or good months. Only once it's clear that the new levels of production need to be sustained is it worth making changes in the number of employees. Things that increase the fixed cost per worker will have some effect, but I would hesitate before blaming Obamacare before I had some rigorous analysis of that question. The first question that I have is how much Obamacare changed the expected cost of hiring an additional employee compared to the cost of finding new employees and training them.

There could be another factor at work. A company might need to expand its physical plant if it hires new workers so it chooses to run the existing machines for an extra hour a day instead buying additional production capacity. Health care is not the only "hurdle" to hiring a new manufacturing employee, there can be an associated capital cost as well.

In my (services) business, if we hire additional people we need to buy them desks and computers and telephones and filing cabinets etc. If we hired more than a few we would have to rent additional premises with a substantial up front cost. In manufacturing, I expect the capital per employee is much higher than the $20,000 or so that we typically have to invest as incremental capital for an average new hire (average capital overall in my business, not counting goodwill or work in progress, appears to be about $80,000 per worker.)

Could risk-aversion be a factor too?

Both upsides as well as downsides are far larger with a new hire. The devil you know.......

Besides Obama Care, this second Obama Administration is likely to be even more aggressive about discrimination lawsuits (e.g., Thomas Perez is being promoted to the Cabinet for all the good work he did suing employers for discrimination at the Justice Department). The more aggressive the feds are in suing over employment discrimination, the more it's in the interest of employers to have fewer but more trustworthy workers.

I have a question, which I'm sure will turn out to be quite naive once someone provides the answer. If in a competitive labor market wages equal marginal product of labor, then why does overtime pay exist? Are workers 1.5 times more productive in their 41st hour of work than in their first 40 hours of the week? Are they twice as productive on weekends and holidays than during regular weekday business hours?

I have thought about it a little bit more and will try to answer my own question. Employers do not bring in overtime workers randomly. They do so during periods of high demand. Thus, even if workers' productivity is constant in, say, widgets produced per hour, it may indeed increase in terms of *dollars* per hour during overtime periods due to increased demand for widgets pushing up the value of widgets. In other words, employers bring in overtime workers precisely during those periods when increased demand for widgets raises their value by 1.5X or 2X.

That does lend support to TC's hypothesis of labor market polarization. If demand for widgets increases slightly, but not by enough to increase their value by 1.5-2X, then employers could respond by hiring more workers at non-overtime wages to meet the demand. The fact that they are waiting to increase production until widget-demand increases by enough to justify overtime pay for existing workers suggests that there were not enough qualified new (non-overtime) workers to hire to have met lower levels of increased demand.

TC is emotionally committed the labor market polarization thesis. This much is obvious.

Another question: if employers would rather pay an existing worker 1.5-2x higher wages for overtime work than hire a new worker at non-overtime wage, does that mean that a minimum wage increase is more likely to prevent unemployed workers from finding jobs than to cause existing workers to be laid off? Would the ratio of long-term unemployed to short-term unemployed also be expected to increase?

Lies, damned lies and statistics. These figures are obviously fabricated by the left liberal conspiracy: it is not possible for any economic measure to improve while a Democrat is in the White House.

We need for those nice Democrat workers to refuse the overtime.

Comments for this post are closed