Unexpected unintended consequences

From Perfectly Nice Policies, With Less-Nice Side Effects by Megan McArdle.

What happens when you suddenly offer parents generous family leave benefits, paid at the expense of the government? You can probably think of dozens of outcomes. But here’s one you might not have been expecting: people die.

That’s the finding of Benjamin Friedrich and Martin Hackmann, in a new working paper at the National Bureau of Economic Research. The culprit? Nurses, who skew female, provide a lot of vital health care, and made heavy use of Denmark’s new paid family leave benefit when it passed in 1994. Since the supply of nurses was limited, and their skills could not easily be replaced, hospital readmissions went up, and more troublingly, mortality spiked among elderly patients in nursing homes.

What McArdle is really focusing on is a real life illustration of Hayek's knowledge problem. A policy maker may have estimable goals in mind and may even have done a reasonable root cause analysis and formulated reasonable policies. None-the-less, regardless of how noble and how intelligent, it is simply not feasible to accurately forecast first and second order consequences. People die.

McArdle links this to the policy debate about minimum wage laws. She provides one of the clearest explanations of the price elasticity of labor demand.

When we assess policies we’d like to enact, we often talk about labor as if it were an undifferentiated lump, and policies will either affect it, or not. This is approximately the level at which minimum wage debates, for example, are usually conducted. But this is nonsense. A higher minimum wage does not affect all businesses or workers the same; it does not even affect all workers at or near the minimum wage the same, or the businesses that employ them.

Some businesses have affluent customers who can afford to absorb a price increase; as long as everyone is force to pay the new higher wage, the cost will mostly be passed on to those customers. Some businesses don’t have affluent customers who can eat any wage differences, but they do have ways to economize on labor, like installing self-service drink stations, or touch screens to take orders. Those businesses will continue operating, but over the long run, will look for labor-saving devices that will ultimately lower employment. And some businesses will simply not be able to keep operating at the higher wage, because their customers are price-sensitive, or they compete with businesses without the same wage requirements -- like an independent bookstore just barely fending off the Amazon threat, or a factory exposed to competition from China. Those businesses will have to close if you raise their labor cost.

Think of the City Council person considering whether to increase minimum wage from $10 an hour to $15. What do you need to know in order to anticipate what the likely consequences of such an increase. Regrettably, most conversations focus on fairness of the "We need a living wage" sort. But say you have an intelligent council person (I know, I know - economists and their unrealistic assumptions.) What do they need to know?

1) How many enterprises are there with employees being paid between 10 and 15 dollars an hour? Those companies are going to be directly affected. This is a first order impact.

2) How many companies are there with employees being paid between 15 and, say, 30 dollars an hour? These companies are indirectly affected. Why? Salary compression? When the minimum wage was $10, an employee with some years experience might be earning $15, a 50% premium based on their relative productivity. If the minimum wage moves to $15, you can't leave the $15 an hour employee at $15, you have to raise their compensation at least to some degree to maintain the distinction between higher and lower productivity employees. This is a second order impact.

3) What are the price elasticities for labor demand for all enterprises affected by the increase in labor cost? If you can't get for individual companies (and you can't), what are the number of companies in each category (and employees affected) and what are the estimated price elasticities of those categories as outlined by McArdle - A) Companies who can easily pass along labor cost increases, B) Companies who cannot pass along increases but have labor capital substitution opportunities, 3) Companies who have no capacity to pass along labor cost increases and suffer reduced profits, and 4) Companies who have no capacity to pass along labor cost increases and insufficient profitability to remain in business. This is a third order effect.

4) What are the operational impacts of labor force churn arising from such an increase? People in surrounding counties where the minimum wage might be $7.50 an hour decide to move into the city to seek the higher minimum wage of $15. Increased densification, rent increases, traffic, etc. These are fourth order effects.

And on and on. Most arguments on minimum wage are emotional and don't move beyond that. Some take a small peek at the possible consequence of the direct effect of the increase, but not often. I am unaware of any instance where sophisticated modeling seeking to address second, third and fourth order consequences is done. But all those effects are real and material.

Minimum wage is simply an example. The main point is that forecasting complex systems is exceptionally difficult and that second order and beyond effects are real and consequential and can, and usually do, swamp whatever the desired first order effects might have been.