No, that's global. If global hashrate goes up, blocks are found faster, thus drop shares increase regarding your hashing output. The size of a single pool doesn't influence that, since you'll drop a share once a block is found anywhere else in the network, for the pool has to build up a new block header.
Oh. I wasn't aware that new work starts on a block whenever anyone finds a block on the network. That makes way more sense than how I was thinking about it. Thanks for the info. Maybe the OP's client doesn't support long-polling, then?