Don Boudreaux has been arguing that monopsony cannot be an important factor in unskilled labor markets because if it were, there would be profit opportunities from starting a new business that employs unskilled labor. If there is no monopsony in those markets, then a minimum wage would tend to reduce the employment of unskilled labor. Since destroying the jobs of unskilled workers is a bad thing, Boudreaux considers this a good argument against the minimum wage.
I don't favor a minimum wage and I think Boudreaux's argument has an element of truth. However, when reading it, I began to think about monopolistic competition. In long run equilibrium, there is no profit and no incentive for further entry. Price is equal to average cost. But price still exceeds marginal cost. The downward sloping demand curve for each firm results in a marginal revenue less than price. The profit maximum, which happens to be a break even point, is where marginal revenue equals marginal cost. Price and average cost are both greater than marginal cost. With a U-shaped long run cost curve, that makes output less than the level that would minimize long run average costs. Each firm has "excess capacity."
Can the same kind of reasoning apply to monopsony? The firms face an upward sloping supply of labor function. If they pay all of their workers more, then more workers will be willing to work for the firm. This makes the marginal cost of labor greater than the wage. The increase in wages that must be paid to the additional worker is an added cost, but the added wages paid to all of those workers who would have being willing to work at the existing wage is an added cost as well.
The firm maximizes profit where the marginal revenue product of labor is equal to the marginal cost of labor. The marginal cost of labor is greater than the wage. The marginal revenue product of labor is also greater than the wage. This implies that hiring another worker at the existing wage would add to profit. But paying more to all the workers so that another worker will come will reduce profit.
This might suggest that entering the industry would be profitable. However, profits depend on revenue and total cost, or equivalently, price and average cost. Does the fact that the wage is less than the marginal cost of labor and the marginal revenue product of labor, have the implication that average cost is less than price? The wage is something that impacts average cost.
Still, consider the following simple scenario. Workers walk to their jobs. The firms are distributed across the landscape. At a low wage, only the workers closest to a firm will choose to work there. At a higher wage, workers a bit further off will work at the firm. Each firm faces an upward sloping supply curve for labor due to geography. This results in monopsony.
Suppose all the firms are profitable. Entry occurs. The firms are bit more crowded across the landscape. The typical worker is closer to a firm. Still, each firm faces an upward sloping supply curve, and maximizes profit where the marginal revenue product of labor equals the marginal cost of labor. However, entry continues until the price is equal to average cost and there are no more profits.
As with the simple monopolistic competition model, there are "too many" firms. Here, each firm is paying too little rather than charging too much. The offsetting benefit of monopolistic competition, the added variety of products, is here represented by having more households having more convenient places of employment--closer to home. The same would be possible with monopolistic competition, more firms would make it possible for more households to have shops closer to home.
A minimum wage, then, would result in fewer firms, each operating at a larger, more efficient scale, hiring more workers and paying them more. Shifting to this new equilibrium would involve transitional losses for all firms and failure for some. The stronger firms survive, face less competition, and are able to expand to a more efficient scale. Some of the workers have to travel further to work because there are fewer places of employment.
Anyway, it seems possible.
As I have explained before, there is a range of minimum wages that would expand employment in a monopsonistic firm. A minimum wage at the going monopsony wage leaves a shortage. Actual employment is supply constrained. As the minimum wage rises, quantity of labor supplied rises, and firms hire the extra workers. At some point a maximum employment is reached and there is no shortage of labor. The minimum wage equals the marginal revenue product of labor. Increasing the minimum wage above that level results in the firm reducing its desired hires. Now employment is demand constrained. At some point employment matches what would have been provided by the monopsonist, though at a lower wage. Of course, there is a much larger range, going to infinitely high, that would reduce employment from what the wage-setting, profit maximizing monopsonist would choose.
Having the political system set a different minimum wage for each firm is unrealistic. Setting a single minimum wage for all firms will almost certainly result in a minimum wage that raises employment in some monopsonistic firms and lowers it in others. And, of course, those firms where monopsony is irrelevant, or even appears irrelevant to the employers, would have a straightforward negative employment effect due to the minimum wage.
Interestingly, a monopsonistic labor market is in shortage. The marginal revenue product of labor is greater than the wage. This creates a motivation for firms to keep their employees happy. When employees leave, employers are anxious to replace them.
Further, if the supply of labor increases, there is an immediate incentive to take on more workers. They will add to profit!
Of course, the firms will have an incentive to lower wages in response to an increase in the supply of labor. Now there is an interesting empirical test. Do firms that receive a resume respond with a new compensation schedule of lower pay for everyone?
If the demand for labor is rising faster than the supply, the equilibrium monoponistic wage will be rising too. And so, a more rapid increase in the supply of labor would just result in slower wage increases. Which is, of course, exactly what would be observed in a competitive labor market.
That is the first test I would use when considering a minimum wage. Is the market in question in shortage? Are all those workers who want to work employed? Do those workers not employed say that the problem is excessively low wages and they would rather stay home? If a labor market is demand constrained, with unemployed workers desiring a job at the going wage, monopsony is not likely to be a problem.