Hyper-threading is about improving local system performance, and has little to do with network throughput. Of course, a well-functioning CPU may be able to push more bits into your network, but that's not usually the limiting factor. That said, hyperthreading can produce a real CPU performance benefit on your systems; I've read that this performance increase runs between 15 and 45 percent, depending on the workload and who you listen to.
Here's where it gets tricky. Even on your local system, most of the time you don't notice this performance benefit. This is for two reasons. The first is that the benefit comes in the form of virtual additional processors, and many local tasks just aren't well-suited to running over multiple CPUs. Even those tasks that are suited for this need to be coded correctly to take advantage of it. The second reason is that the dirty secret of desktop performance today is that it's not about the CPU any more; it's about the hard disk. So even if you have a system that would take advantage of hyperthreading nearer the 45% mark than the 15% mark, you'll still probably get a lot more bang for your buck ensuring that you have good hard disks and enough RAM. Where I believe hyperthreading really shines is in those situation where you sometimes have a process with several threads that runs away with your CPU. Hyperthreading can help ensure you always have available CPU time for normal desktop tasks.
The upshot is that there's a lot of benefit to be had for relatively small cost, especially compared to the cost of your employee's time, from providing well-specced computers. For this specific model, you only need to save less than a minute per day of your employee time over the life of those CPUs to get savings that exceed their cost difference, even at minimum wage1.
1Cost difference of $80 spread over and estimated 4 years and ~220 work-days per year is about 9.1¢ per day. At $7.25/hr it takes about 45 seconds to earn 9¢