Both my PC and Switch have shielded jacks on them, and those have continuity to the chassis ground.
In fact, I just searched Amazon for 1g and 10G switches, and they all have shielded jacks on them.
Now... if someone chooses to use unshielded patch cables, you have to live with the possible lower performance, as I suspect is specified by the standards.
Yes that's normal for gear to have shielding - it's as much to provide continuity for shielded cabling as it is to shield the signal from the noise made by the device itself (or the device from neighboring devices). The possible lower performance is really a non-issue, because the cable's analog bandwidth is so much in excess of what's required for digital data transmission that 100MHz (or less of performance difference just isn't relevant.
for example - here's some panduit cat6 cable:
http://www1.panduit.com/heiler/CatalogCutSheets/PUR6004BU-UY Product Page.pdf
UTP - certified to 550 MHz. 250Mhz is what you need to be rated for gigabit. That's 300 MHz of bandwidth beyond what the standard requires. even if it's done by a hack, and they wire it wrong, 9 times out of 10 it will still pass bandwidth testing. I know because I was the guy with the $10k+ Fluke meters certifying cable for 5 years - state orgs, private companies, private datacenters, etc. I almost never saw shielded cabling, to the point were I was the only one who did it when it was called for.
One of my favorites was the non-contiguous "foil" layer in someone's cabling, because it rejected a lot of crosstalk but was non-conductive down the cable, and all it required was trimming. got good results and wasn't hard for guys to handle.
sometimes you actually get reject 6A supplied as cat6, because it failed/was marginal on testing so they slap cat6 on the jacket and sell it. I've never bought name brand (panduit, systemax, belden, etc) cable/jacks that failed to pass with flying colors. I have seen ebay wire/jacks be extremely marginal (not to mention hard on the hands to work with, poor finish, casting/mold flash, etc).
Those little dividers in the cabling (X or H shape, or a ribbon) make a big difference. usually the interference you're looking to escape isn't outside sources, but from the other pairs themselves. They put a lot of tricks into the insulation (even offsetting the wire from the center of the insulation on individual pairs), to get lower crosstalk out of it.
The environment I would be looking for shielded cabling in would be an industrial plant with lots of high frequency electrical equipment (e.g. manufacturing plant with lots of VFDs) with ethernet wire run right up to/inside the gear. this is where shielding starts to shine. otherwise it's generally not worth the trouble or the expense.
