Bitcoin is of course not an ideal transport layer, but it is certainly workable. The 'no SPV clients' issue, for instance, may be easily avoided by asking for consensus from a federation of servers together trusted to report accurate balances, and lots of people use full-blockchain clients anyway, even though the time to initialise Bitcoind is many times longer than the time to initialise counterpartyd after that. I'm quite sure that the benefits of building on Bitcoin, not the least of which are much increased development speed and security, do indeed outweigh the costs.