This is why the National Restaurant Association always lobbies furiously against minimum wage hikes - it will never destroy the restaurant industry but it sure as heck destroys actual restaurants.

Which seems absolutely crazy to me. I would have expected them to push for them harder than your typical org since such a sweeping change would affect everyone basically the same. Nobody would be immune from raising their prices and the value differential would nudge customers away from fast food.

I think it doesn't affect not-yet-open restaurants nearly as badly as existing restaurant owners. NRA doesn't represent the restaurant market as a whole it represents a % of restaurant owners that currently exist.

Restaurant workers are, of course, going to be fairly indifferent to restaurant bankruptcies especially if an existing restaurant is swiftly replaced by a new one that pays more.

You might think that, but errors and fluctuations tend to routinely cause high margin businesses such as software to collapse in large numbers. Think of the dotcom bust. I suspect part of the problem is these businesses flourish when the conditions are easy, but many don't have the experience or exceptional margins to survive long enough to adapt when conditions change.

Perceived high-potential-margin businesses with lead time lead to speculative investment in low- (or negative-) margin businesses in the same field with perceived potential, which is sensitive to fluctuations in the broader investment climate (e.g., payoff timelines) or reassessment of profitability potential. That’s what the dotcom boom and bust were about.

The actual high-margin businesses wouldn't be taken out in the bust unless they were high margin only because they were selling into the boom.

"The actual high-margin businesses wouldn't be taken out in the bust unless they were high margin only because they were selling into the boom."

Exactly.

The dot com bust was negative margin businesses.

When I look at Uber I don't think anything has changed.

Some businesses are inherently more controllable: for example, you cant expect 5% accuracy on NASA building a telescope that has never been done before but you can expect +% accuracy on the price of commodity product