On the other hand PhD applicants are much closer to the intended purposes of the 457. For example research cannot be developed in the same manner and quantity as IT.
That's a bold statement. Have you got some good source? I'm having a problem imagining a statistically relevant number of people saying "I'm interested in pursuing one of the skills marked by the government as experiencing national shortage, (for years!) but I won't because immigrants will take those positions instead"
- Relative size of CS departments (enrolments) and how standalone they are as a faculty now vs 10-15 years ago
- Observations of how many parents who work in IT recommend to their kids to pursue alternative fields of study
Statistically I am not sure how this fits in with other non-IT fields.
Personally I would love to see the points above greatly improved upon as computers are here to stay, so having self-sufficiency and cross-generational (family) expertise would be very complementary and competitive.
However we don't have an infinite supply of jobs and every job taken by a skilled person from another country could have been taken by a skilled person from within our own country, had they developed the skill. It also reduces demand for those jobs, which reduces the motivation for people within the country to train themselves to perform it in the future and increases our dependence on immigration.
My argument isn't that immigration is bad, I'm an immigrant myself (Australia -> US) but it's definitely a trade off and needs to be controlled carefully.