But if a young adult cannot produce enough of value to justify being paid a living wage, nothing we do to the minimum wage will help. He, the institutions which trained him and the society in which he lives, have far bigger problemsHarford's certainly right that the longer term solution to low pay is productivity increases; you can't mandate paying people more than their marginal product as firms will simply shift to more capital-intensive processes or to solutions through offshore outsourcing of labour-intensive components.
But not all wages need to be living wages. The 16 year old living at home working part time while going to school doesn't need one; and, his marginal product probably isn't enough to finance a real living. But the kid is learning valuable skills about workplace culture, showing up on time, dealing with customers - things that will help him be more productive in future. Note further that the minimum wage only counts pecuniary benefits; health insurance that comes bundled with many, but not all, US jobs is an increasing proportion of the overall salary package.
At least, as Harford notes, the UK Low Pay Commission, which advises on minimum wages, has let youth minimum wages stay relatively low. Workers aged 21+ are on a minimum wage of £6.08; workers aged 18-20 on £4.98; 16-17 year olds on £3.68; and, there's also an apprentice rate of £2.60.
Here in New Zealand, National's making it a bit to get younger folks onto the lower New Entrants' Wage; otherwise you have to pay the 16 year old, on his first day of work, no less than you'd pay any other minimum wage worker: $13/hour, (US $10.30 or £6.73).* Our youth unemployment rate isn't pretty, at least partially as consequence of the prior Labour government's decision to bring 16 and 17 year olds up to the adult minimum wage.
We can mandate that all wages are living wages, but we can't mandate that all the people who'd like to have work at that pay are able to find jobs.
And I worry Harford packs a bit into "market power" here when explaining Card & Krueger's results on minimum wages and employment in New Jersey:
If employers have market power in the labour market then they might actually offer a lower wage than the balance of competitive supply and demand would produce. Some workers would rather keep looking or sign up for welfare payments, and so employment is lower at this level. Introduce a minimum wage and both wages and employment increase, while profits fall.Imagine a remote mining town with one big employer who pays the same basic wage to everybody. If you can't pay the new guy more than your existing workers, then you have incentive to abstain from hiring somebody whose output would cover his salary but wouldn't cover the amount you'd have to pay in salary increases to all your other workers. So increases in minimum wages can increase employment and wages in places with monopsonistic employers and low labour market mobility. But in low-skilled retail in suburban New Jersey where there's no way that hiring an additional student to stock shelves at the 7-11 forces you to push up your other workers' wages? Any retailer who's stuck with a labour shortage at current wage rates and the chance to hire an additional worker for a wage less than the worker's marginal product but above prevailing rates has incentive to defect from any cartel of employers. And remember that the theoretical argument works by having people enter the workforce in pursuit of the new higher minimum wage who were outside of the workforce previously; if we start with unemployment - people in the workforce who cannot get jobs at the prevailing wage - the theory can't really apply. I'm not sure that many of our current problems come from too many folks voluntarily sitting outside the labour force because prevailing wages are too low.
* At last report, median hourly earnings were $20.38 (average $24.78). Our minimum wage is 63% of the median wage or 52% of the average; at these levels, we expect reasonable disemployment effects, especially among youths.