As unemployment falls, the marginal product of labour falls. That is, each succeeding person hired becomes progressively less suitable for the vacancy they fill. This occurs because the fewer unemployed there are to choose from, the less likely an employer is to be able to find the perfect employee to fill each vacancy.
When this “product” or output of those hired falls to the minimum wage, union wage, etc etc, unemployment cannot be reduced any further, else employers tend to resort to poaching each other’s employees (consciously or not) rather than take labour from the dole queue, thus inflation kicks in.
So what’s the solution? It’s easy – in principle. Just subsidise those marginal employees.
Identifying the marginal employees is easy at first: they are the unemployed. But there is a problem: as time passes, the unemployed who have been subsidised into work are no longer “marginal”, i.e. they are not the least productive people in each firm.
This is for several reasons. First they gain firm specific skills and begin to become economic propositions even without any subsidy. Second, the pattern of supply and demand for different types of labour on each local labour market, and within each firm, is constantly changing: types of labour which were in surplus, and which were thus “marginal” one month, may not be in surplus six months later.
So how do we ensure that employers claim the subsidy only in respect of those employees who are genuinely marginal? Easy: call the employer’s bluff. That is, limit the time each subsidised employee stays with a particular employer, or remove such employees from subsidised jobs at a random interval after they start their subsidised job. Employers DO MIND losing valuable employees, but they DON’T MIND losing peripheral or marginal employees.
There are of course a number of refinements or additional rules that could be added to the game. For example an obvious way for employers to cheat would be claim the subsidy in respect of employees who were not marginal (i.e. who were economic propositions without the subsidy). And then when the subsidy expired for a given employee, the employer could simply continue employing the employee on a normal or unsubsidised basis. The solution to that problem is to bar employees from returning to a given employer for several months if the subsidy expires before the employer has voluntarily ceased to claim the subsidy.
A disadvantage of the above system is that it would involve bureaucractic expense. But then half the problems of the labour market stem from artificially imposed bureaucratic interferences: union wages, minimum wages, unemployment benefits, etc. These may be worthy “interferences” but they constitute a move towards having labour allocated by the bureaucracy rather than by the market.
So there is a choice: a genuinely free labour market, or a bureaucratically run / regulated labour market. If you want the latter, don’t complain about bureaucratic expense.
Another disadvantage is that turnover amongst the least skilled and talented would rise. But then many people quite happily work for temporary employment agencies, and that often involves several “jobs” a week. So the “high turnover” point is a weak objection.
The Edmund Phelps subsidy.
Phelps* advocated a subsidy of ALL low paid labour. The disadvantages of that idea relative to the above “marginal” idea as follows.
1. The total number of people subsidised under the Phelps subsidy would be far more than under the marginal version, thus taxpayer costs be higher. But those costs should NOT be seen as consisting entirely of REAL or RESOURCE costs.
2. The Phelps subsidy involves subsidising many people who would have been employed anyway: that reduces the incentive to get work out of such employees, and that certainly is a REAL cost.
3. The marginal idea gives the option of taking the marginal product of labour down to zero or near zero, which might seem a silly objective. But it could be justified, first, in “work experience” or training grounds. Second, it could be justified on what might be called “workfare” grounds. That is, one way of calling the bluff of unemployment benefit spongers is to tell them to turn up at some place of work (even if they don’t do very much) else their benefit gets cut.
Third, the fact that no one is prepared to pay anything for a product does not prove the product is worthless. That is, given excess unemployment, the reason people are not prepared to pay for extra units of all the products they already consume is not that they regard those extra units as worthless: if aggregate demand were to rise, those extra units WOULD be purchased. So why not so to speak just produce a few extra units of everything and distribute the extra units to everyone – which is more or less what zero revenue product labour achieves under the above marginal idea.
* Phelps, E. S. (1997), Rewarding Work How to Restore Participation and Self-Support to Free Enterprise, Harvard University Press.