A few points to make:
First, although I agree with you (for the most part) that perceived prestige is important, people who are receiving and reviewing applications and interviewing applicants are well aware of the University of Washington's high prestige as a research institution (in general) and as a clinical psychology program (specifically). If they're looking for prestige, they've got it with an applicant who's studied under Linehan. Although Harvard is a big, beautiful name, the program itself was only accredited recently (last summer? correct me if I'm wrong) and has some great researchers, but not quite the same reputation- yet.
I also think that looking merely at percentages and not the overall numbers is misleading. Harvard placed 100%, but out of 16 people in 10 years. This is probably because it's a fledging program, and my guess is that if you compare it to other fledging programs (without the big, beautiful name, see, maybe Seattle Pacific: approx 50%) the numbers would go down. Yale's more established, but the program still placed 34/34 in 10 years. Very impressive, but not a great output for a program that take 4-6 students per year (
http://www.yale.edu/psychology/clinical_perfdata.html). Where are all the other students going? Nowhere.
On the other hand, UW placed 88.9% (an excellent percentage, BTW, I'm not sure where you're getting the "only" thing) out of 99 students. It's certainly true they didn't place every student (I'd be shocked though if those numbers didn't go up to at least 95-98 after clearinghouse) but it's a large enough sample to actually consider the issue in the first place, and they're moving students through the program
reasonably fast. (Although the number of people taking 7 years alarms me a tad, I'll admit (
http://web.psych.washington.edu/areas/admissionsdata/clinical.html). But keep in mind, when one person out of ten doesn't match, the percentage goes from 100% to 90%- like that. Whether or not they later match in clearinghouse.
The math on all this can look much more... prestigious than it is. Without all that, I'm not sure your base argument makes much sense- the internship match isn't (it doesn't seem to me) about prestige for the sites, it is about quality applicants who fit, period. Is it shocking an institution that spends a lot of its money investing in resources, researchers and mentors for its students would be well-equipped to train such an applicant?