In a competitive labor market, raising the minimum wage results in lower employment, lower output and higher prices.
But we just saw "How Silicon Valley CEOs conspired to drive down tech engineer wages," so we know we don't have a competitive labor market. See https://news.ycombinator.com/item?id=7111531
In a monopsonistic labor market, raising the minimum wage results in greater employment, greater output, and lower prices. It results in greater employment because the monopsonist faces and upward sloping supply curve for labor, and uses his market power to keep wages artificially low, beneath the equilibrium wage. If the wage is exogenously raised, the quantity of labor supplied increases. (If you don't believe me, you should draw supply and demand curves, and prove it to yourself.)
In reality, when NJ raised its minimum wage in 1992, what happened was (1) employment increased, and (2) prices increased. This observation fits neither the competitive nor the monopsonist model! These observations were made by David Card and Andrew Krueger.
A resolution for the "Card-Krueger Paradox" was developed by A. Ross Shepherd, Professor Emeritus of the University of Missouri--Kansas City and Published in the Southern Economic Journal on Oct 1, 2000. There's a copy online here (but it lacks the figures): http://www.thefreelibrary.com/Minimum+Wages+and+the+Card--Kr...
His solution boils down to realizing we have an impure monopsonistic market in which the higher minimum wage increases the firm's Long Run Average Cost. So employment goes up, but so do prices.
Because in the alternate universe where there is no minimum wage, the people who work at McDonalds do not get paid $5 an hour. They belong to a union.