Threat of new entrants

High incumbent returns are an advertisement. If they persist, something must be blocking the response to that advertisement. Your job is to name the something, put a price on it and check it against the actual entry record. That record is the empirical test the narrative can't fake. Here's Porter's barrier inventory (2008 formulation), each barrier with its analyst's test:

1. Supply-side economies of scale. Unit costs fall with volume (plants, purchasing, logistics density). To test it, ask what share of the market an entrant must win to reach cost parity. Is it 2% or 20%? The bigger the minimum efficient scale relative to the market, the fewer firms can play.
2. Demand-side benefits of scale (network effects). Buyers value what other buyers use (platforms, standards, marketplaces), which are the archetypes you met when naming a company's charging machine. Test it by asking whether willingness-to-pay rises with the installed base.
3. Customer switching costs. Retraining, re-integration, re-certification, data migration. Test it by working out what a customer actually spends (money, time, risk) to change vendors. And look past the end user, because an intermediary can hold switching costs too, like the paint dealer's tinting machine or the pharmacist's habit.
4. Capital requirements. This is the most over-rated barrier on the list, and Porter says so. Capital alone deters only when it must be sunk into unrecoverable, entry-specific assets, such as a brand launch or R&D. Financeable capacity works as an invitation to enter, as you'll see when the capital cycle comes up. Test it by asking whether the required capital is recoverable on exit. If it is, discount this barrier heavily.
5. Incumbency advantages independent of size. Proprietary technology, best locations taken, raw-material access, learning-curve depth, brand built by decades of accumulated (sunk) advertising. Test it by asking whether money could replicate it in five years. If it could, you're looking at a head start and no barrier at all.
6. Unequal access to distribution. Shelf space is finite, and dealer loyalty compounds. A new consumer brand must buy its way into channels incumbents got free. To test it, count the doors (outlets, dealers, distributors) and estimate the years and cost to replicate them. E-commerce and D2C have lowered this barrier in consumer categories, so check whether the incumbent's channel still controls the customer.
7. Restrictive government policy. Licenses, spectrum, banking permits, drug approvals, FDI caps, tariffs. When it's present it's the cleanest barrier of all, and the easiest to reverse with a stroke of a pen.

All seven get multiplied by one more thing: expected retaliation. Entrants read history. An incumbent with deep pockets, high fixed costs to defend and a record of scorched-earth response (price cuts, channel lockouts, poaching) will deter entry before it happens. To test it, ask what happened to the last entrants. Name them and their outcomes. That one table is worth more than any theory.

What evidence shows how high the barriers are?

For entry, the evidence comes from these places.

For anchors, an entry threat of 1 is a multi-decade barrier stack being tested and holding, with entrants' cumulative share stuck in low single digits. An entry threat of 5 is this year's franchise, open to anyone with a lease and a loan (restaurants, generic trading, most e-commerce niches).