The Minimum Wage Is Like A Proxy Labor Union

In his State of the Union speech Barack Obama warmed Democratic hearts by calling for an increase in the federal minimum wage to $9 an hour, and indexation of the wage to inflation.

Predictably, an op-ed battle immediately sprang up, with half of the commentariat arguing that a higher minimum wage would simply hurt low-skill workers seeking a job (and some adding that wage subsidies would be a far better option) and the other half arguing that higher wages at the bottom of the income spectrum are long overdue.

Ezra Klein falls into the latter camp, saying that the chart below should tell us everything we need to know about the case for a minimum wage:

Mr Klein argues that a minimum wage is like a proxy labour union; sure it may have some employment effects, but it effectively raises the wage bargaining power of those workers who do manage to find employment. In the absence of such bargaining power, we can't expect any meaningful increase in wages at the low end of the income spectrum.

What should low-skill workers themselves want? The empirics of the minimum wage debate are complex, but it isn't impossible to draw some conclusions. A recent Free exchange column attempted to do just that:

In sum, the employment effects we'd expect if labour-markets were perfectly competitive don't emerge. That's because there is some monopsony power to labour markets, associated with frictions like the cost of searching for new jobs. Those frictions give employers a bargaining-power advantage that a minimum wage can in some cases counteract.

Further, minimum wage increases may give both workers and employers an incentive to raise their productivity levels in order to preserve jobs: people work harder to justify the higher wage. That helps explain why minimum wage increases can influence pay higher up the income scale.

That dynamic—that higher minimum wages often have less of an employment effect because they cause workers to exert more effort—is also something that left-leaning supporters of higher minimum wages should take into account.

One thing that should seem clear is that the search frictions that give rise to low levels of employer monopsony power are not at all sufficient to explain the extraordinary surge in capital's income share.

For that we probably need to point to technology, which has substantially increased labour-market competition for many low- and medium-skill tasks. Interesting new economic work is modeling a world in which offshoring and automation are often substitutes, and a firm's choice to employ local workers, offshore, or automate is determined by both relative productivities and relative costs.

The implication of this work is that some labour-market segments are likely to show more of an employment effect than others. In occupations consisting of mostly routine work, it will be very easy for an employer to respond to higher costs by moving the job elsewhere or swapping in a machine. Think of check-out clerks.

While many retailers have invested substantially in automated check-out, others have not, presumably because the cost difference between human and robot is not yet big enough to justify the change. It seems very probable that higher minimum wages will in some cases tip that balance.

In other segments, consisting of much less routine work, one would expect less of an employment effect—maybe none, for modest increases. That doesn't necessarily mean that the minimum wage increase would be entirely benign, however.

It's possible that displacement from routine-task occupations could increase competition for non-routine-task occupations. That may push down wages in some cases (where wages remain above the minimum wage), and raise joblessness in others.

The latest research suggests that a carefully imposed minimum wage (and I would consider Britain's carefully imposed, and one indexed to inflation carelessly imposed) can raise incomes at the bottom of the wage spectrum without much reducing employment.

But a higher minimum wage is neither a sufficient or a particularly germane response to labour-market polarisation and capital's rising income share. The problem is simply that the supply of people and robots available to do routine work is exploding.

A proper response to this dynamic must either be a big change in relative skill supplies or relative productivities, or a move toward wage subsidies that are far larger and broader than have been considered in the past.

Click here to subscribe to The Economist