I don't think broccoli would matter unless it were already the cheapest vegetable with no substitute good and common to the cuisine of specific person. But if

I don't think broccoli would matter unless it were already the cheapest vegetable with no substitute good and common to the cuisine of specific person. But if you're trying to generalize the idea that increasing minimum wage would lead to an inflation death spiral because labor is too high a percentage of cost of staple goods... I'm not sure I buy it, though agriculture is definitely one of the industries most impacted by low cost labor (I don't know what they pay the folks crossing the border at 3 AM to pick lettuce in Yuma in 100+ degree heat, but it's not a lot). But my general sense is that the labor to harvest isn't most of the component cost of lettuce, there's transport, refrigeration, irrigation, pesticides, etc., so you can probably assume doubling minimum wage doesn't do anything like double the price of lettuce. If you believe that the labor to pick lettuce is 50% of the cost of lettuce, your worst case for a 2x wage increase is a 1.5x gain in the price of lettuce, which means the low end of the economy is getting net purchasing power growth, but more realistically, I can't imagine that picking lettuce is 50% of the cost, meaning low-end labor wage increases can't have near that much impact on prices. My assumption is that the issue with higher labor costs is foreign competition at lower cost with lower standards causing a competitive issue and net less business more than it's an "wages increasing the price and causing an inflation issue." (IE the general "US must have 3rd world wages because otherwise it can't compete with the 3rd world" idea.) But in my view so long as the increase in minimum wage creates an increase in overall purchasing power greater than any follow-on cost of inflation, it's probably net positive, and should be more efficient because the more normalized wages and living conditions are circa median income, the greater benefit to economies of scale and services and such for that group. Having too many poor and too many distinct socioeconomic bands is bad for the market because it means that the next model-T type product will be more likely to fail. So dealing with the issue caused by competing with lower-wage producers outside our boarders should be addressed separately. (For example, subsidizing US labor's wages by how deficient the living standards of competing labor is, or some other, smarter, plan, someone smarter than me comes up with.)