It seems to me that the answer would be to have a pilot program in some number of stocks that are scattered across different average volumes and market caps, in which the specialist acts as auctioneer/co-ordinator only, and cannot trade for his own account at all. A second group of stocks could be traded on the ECNs only, and not on the floor at all. The results can then be studied in comparison to similar stocks not in the programs and a determination can be made as to which of them makes sense. Including stocks that report earnings during the test would be a good way to identify the strengths and weaknesses of the different options.
AIG's contention that the specialist should take a longer-term position in the stock seems ridiculous to me. Nobody ever said the spec was supposed to smooth out multi-day fluctuations in a stock - their activity is supposed to be very short-term only (minutes/hours).
The comparisons with the stock markets of other countries or markets for other products don't make a lot of sense, since they are much lower volume or have other different trading characteristics than NYSE stocks.
Personally, I think the answer is to leave the system as it is, and simply enforce the existing rules.
AIG's contention that the specialist should take a longer-term position in the stock seems ridiculous to me. Nobody ever said the spec was supposed to smooth out multi-day fluctuations in a stock - their activity is supposed to be very short-term only (minutes/hours).
The comparisons with the stock markets of other countries or markets for other products don't make a lot of sense, since they are much lower volume or have other different trading characteristics than NYSE stocks.
Personally, I think the answer is to leave the system as it is, and simply enforce the existing rules.