To begin, I don't understand why the argument that spoofers have the risk of getting their orders hit (bluffs called) by aggressive traders necessarily makes their practice anymore legitimate. A pump and dump scheme comme Wolf of Wall Street has risk; a short seller could call their bluff and blow it up in their face. Does that make pump and dump schemes acceptable? Just because they have risk doesn't make a practice legitimate. Mexican cartels have risk.
I think we can all agree that price changes are a function of order flow, so if you agree with the above notion that front running is the act of knowingly entering the market before large order flow, then you should agree that all traders are attempting to front run - HFTs, human beings day trading, and buy and hold investors alike.
Taking this logic further, if spoofing indeed misleads those attempting to front run, spoofing then affects all market participants negatively.
If your argument was that "we want to curtail HFT, so we suggest allowing spoofing orders to persist for 1 millisecond" or some time horizon that only would be actionable by HFTs, then you might have a plausible argument for why allowing spoofing would lead to less of the front running that folks on this thread and Arnold seem upset about (i.e. HFTs picking off funds rebalancing, iceberg orders, etc.). However in practice, we aren't talking about orders that last 1 ms; we're talking about spoofing orders that stay in the book long enough to mislead a significant number of traders. It takes a sucker for spoofing to work. HFTs are smarter than you're giving them credit for. They adapt to market conditions. If there was an equilibrium of spoofers and institutions trying to disguise flow, HFTs would move away from this approach. If there were more spoofers, they would get in on the opposite side of the spoof like the trader attempting it and try to ride the suckers who implicitly believe that markets move toward size. You aren't going to stop sophisticated traders attempting to take advantage of flow by allowing spoofing; you're just going to introduce more noise to the book and, by extension, the market in my opinion. Remember, everyone is trying to get in front of order flow, whether you're looking at my timeframes or John Arnold's.
I think there's room for an honest discussion about allowing spoofing in some form, but I don't think that allowing a long banned deceptive practice just because it will temporarily marginally disrupt HFTs makes sense. Here's a good blog on the topic.
I think we can all agree that price changes are a function of order flow, so if you agree with the above notion that front running is the act of knowingly entering the market before large order flow, then you should agree that all traders are attempting to front run - HFTs, human beings day trading, and buy and hold investors alike.
Taking this logic further, if spoofing indeed misleads those attempting to front run, spoofing then affects all market participants negatively.
If your argument was that "we want to curtail HFT, so we suggest allowing spoofing orders to persist for 1 millisecond" or some time horizon that only would be actionable by HFTs, then you might have a plausible argument for why allowing spoofing would lead to less of the front running that folks on this thread and Arnold seem upset about (i.e. HFTs picking off funds rebalancing, iceberg orders, etc.). However in practice, we aren't talking about orders that last 1 ms; we're talking about spoofing orders that stay in the book long enough to mislead a significant number of traders. It takes a sucker for spoofing to work. HFTs are smarter than you're giving them credit for. They adapt to market conditions. If there was an equilibrium of spoofers and institutions trying to disguise flow, HFTs would move away from this approach. If there were more spoofers, they would get in on the opposite side of the spoof like the trader attempting it and try to ride the suckers who implicitly believe that markets move toward size. You aren't going to stop sophisticated traders attempting to take advantage of flow by allowing spoofing; you're just going to introduce more noise to the book and, by extension, the market in my opinion. Remember, everyone is trying to get in front of order flow, whether you're looking at my timeframes or John Arnold's.
I think there's room for an honest discussion about allowing spoofing in some form, but I don't think that allowing a long banned deceptive practice just because it will temporarily marginally disrupt HFTs makes sense. Here's a good blog on the topic.