Saturday, March 23, 2019

Forcing Wages Up Is a Bad Deal For Workers

Suppose you own a factory staffed with assembly line workers or a warehouse staffed with runners who stock and retrieve items. How cool should you keep your building? You’re faced with an optimization problem: What is the profit-maximizing temperature? You could totally skimp on climate control and let the building get unbearably hot in the summer. But this doesn’t save you money, because you would then have to pay your workers more to tolerate the poor working conditions. Workers know that they will be uncomfortable, perhaps even miserable, working a physically demanding job in the heat. They presumably are aware of the potential for heat stroke (or they soon will be). They will either quit en masse or they will demand adequate compensation to continue working. Skimping on climate control will cost you money. There is a sweet spot somewhere, a temperature (let’s say it’s 75 degrees) that minimizes the sum of your labor costs and your utility bill.

It’s useful to think about what happens when you try to deviate from this sweet spot. Maybe you run some numbers, and increasing the temperature to 80 degrees saves you 50 cents per employee-hour on utility bills, but you’d have to pay them an extra $1/hour to endure the conditions. Skimping is a losing deal for the employer. Or, to put it another way, you have the opportunity to sell something to your employees that only costs you 50 cents but which they value at $1. (That something being “an hour’s worth of 5-degree temperature mitigation”.) Usually there's a deal to be made when one party can produce something more cheaply than what another party is willing to pay for it.

Employers negotiate these trade-offs all the time, sometimes implicitly and sometimes explicitly. Should a restaurant provide free meals for its employees at break-times? Maybe. (A restaurant I worked for in high school had this policy.) If it doesn’t, that means its employees will have to take a longer break for meals, go somewhere else, and purchase a meal. They’ll demand higher pay, knowing they’ll need to purchase a meal in the middle of their shift. The restaurant might have to compensate its employees something like the market price of a meal if it makes them fend for themselves, but a restaurant can usually produce a meal with a (for example) $10 market price at an expense much less than $10. It costs something to provide free meals; ingredients and line cooks cost money. But the savings on payroll make up for the expense.

Should an office have refrigerators and microwaves for its employees to use? Again, the company could scrimp on these expenses. The appliances cost something, and refrigerators impose a cost on your utility bill. But employees without a fridge and microwaves might need to either eat unappealing “nonperishable” lunches brought from home or go to a restaurant and purchase a meal. It’s worth something to the employee to have the option of relatively cheap home-made meals. It might be cheaper to buy the appliances than to compensate the employees for doing without these amenities.

Should you make employees buy their uniform or provide it for free? Give them a 15-minute break every four hours or make them work through? Give them flexible working hours or insist on strict clock-watching? Free coffee? Showers and a changing room? A gym? A training or mentorship program? Should you implement borderline-compliant standards of safety or above-and-beyond standards of safety? For all of these options, employers and employees negotiate a trade-of. The employer might think of itself as minimizing its total expenses. The employee might think of itself as purchasing an amenity from its employer, an amenity that the employer can produce relatively cheaply. The employees could earn higher wages if they all decided to forego all amenities, but they don’t want to.

Enter the minimum wage. (Or anything that “raises wages” by legislative fiat.) The naïve view is that the minimum wage simply raises all the affected workers’ wages and nothing else happens. No negative consequences whatsoever.  A slightly more sophisticated view, based on empirical work showing a low elasticity of demand, is that there is some minor job loss but the increase in wages more than compensates for this negative side-effect. An even more sophisticated view would account for employers raking back the fringe benefits (and fridge benefits) described above. The employer would have paid its low-skilled workers $7.25 an hour and given them free meals, flexible breaks, and free parking. Oops, now the employer has to pay, say, $10 an hour. The employer doesn’t take this lying down. That business's underlying economics haven’t changed. The marginal productivity of its employees hasn’t changed at all, but the employer now has to compensate them more in wages. It’s going to pull back $2.75 an hour in amenities, wherever it can find it. We know that those now-discontinued amenities were worth at least $2.75/hour, because that’s how much (in this example) the employees were paying for them, and there is almost always some amount of consumer surplus on top of what the consumer actually pays.  Given the loss of consumer surplus on amenities, we’ve almost certainly made these people worse off. But someone naively running regressions on government employment and wage data might wrongly conclude that we’ve helped them.

Running regressions on government employment statistics is exactly what many empirical researchers do. It’s a “Well, here’s the data we have” approach, like the figurative drunk looking for his keys under the street light. I have heard one economist lampoon this approach in the following way: These economists are living up to the straw-man stereotype that they only care about dollars and cents. They don’t seem to care that the work environment might become less pleasant or even intolerable if we force wages up by fiat. They don't seem to care that there are things an employee might value about their job aside from the dollars earned and hours worked.  It doesn’t seem to register with them that there are things they can’t measure. (Wouldn't it be stunning if people didn't care at all about the quality of their working environment, given that they spend a quarter to a half of their waking hours there?) It doesn’t even seem to register with them that their own field meaningfully informs us about those things we can’t measure. The discussion of amenities above is stuff from basic econ 101. It's simply applying the welfare economics you might encounter in a standard microeconomics textbook to the question of how much to pay in wages versus how much to pay in amenities. When we overrule people's agreements and make them take a deal other than they one they've chosen for themselves, we are far more likely to harm them than to help them. Thinking we can help people by forcibly altering their labor contracts is like thinking we can help people by removing and replacing items in their shopping cart.

_________________________________

I should probably leave this part off, but I'm tempted to preempt a certain kind of response. It goes something like: "Yeah, right. Employer can turn up the heat and tell their employees to go suck it!" Perhaps followed by actual examples of Amazon runners suffering heat stroke and Chinese factory workers committing suicide over poor working conditions. (I've spoken to someone who worked at an Amazon fulfillment center, and by his account the media portrayal is a gross exaggeration.) Some people think that employers can claw back fringe benefits at will because they have "market power" or something. Okay. 1) So why do any employees offer any amenities at all to their employees? If employers can engage in arbitrary expense-slashing without incurring these other costs, why not do it? 2) Even assuming employers have market power and can pay lower than the fair market rate, they are still trading off between amenities and hourly pay. They would still have to pay their employees more in wages if they lowered their safety standards (for example), even if the total wage + amenities is lower because of market power. 3) The horrific real-world examples that come to mind, like heat stroke and draconian work protocols in Amazon fulfillment centers, are probably examples of exactly what I'm describing in this post. These examples usually involve low-skilled workers on whom the minimum wage or some other mandatory benefits legislation is binding. The employer legally must pay a higher wage than they otherwise would, so they slash amenities and try to squeeze more productivity out of each worker.

No comments:

Post a Comment