My argument is that it is theoretically possible that in a declining market environment, the combined effect of HFT and other algorithmic traders (I apologize for conflating and using crude terms, but my vocabulary is admittedly imprecise and incomplete) could accelerate a market selloff and overwhelm the normal mechanisms that have evolved over time.
For example, one worry I have is that algorithms may crowd out non-silicon short sellers. As much as short sellers are demonized as "evil speculators" in times of market stress, sometimes in a down market they are in fact the only buyers available and actually slow down declines when they cover. I don't have solid evidence that short sellers are crowded out by HFT... but I am trying to provide an example of a scenario where the "macro" effects could transcend the nuanced "micro" arguments about cost, efficiency, liquidity, etc. There are numerous other plausible scenarios that fit this bill as well, and it is the uncertainty around them that I am trying to point out.
Of course, none of these things have to do with IEX or a time delay... maybe things like order stuffing or spoofing are related but even those things are not what I'm talking about. Especially considering the dialog that has taken place since Flash Boys was released, I feel that both industry and regulators risk getting caught up in a cost/benefit analysis of the small issues while ignoring potential systemic problems.
As an investor I don't really have a huge dog in this fight... I'm not expecting any pension, and I don't own any managed funds. I generally use fundamental analysis to do long term investing in broad sectors using ETFs and a smattering of individual stocks. I manage my own money and so I will have no one to blame but myself if things go wrong. And frankly, if I lose $0.0002 on a trade by being frontrun in a dark pool it doesn't change my life, so for me to argue for or against IEX is somewhat irrelevant. However as a taxpayer I am concerned that public pension systems could very quickly find themselves vastly underfunded in a short period of time, and as a human being I worry about people who have worked their whole lives and just want to retire in peace.
I think that my concerns are not entirely disconnected from what the general public feels/understands about "modern markets", so while in most cases I probably have no basis to refute your specific arguments, I suppose I take exception to people with "macro" views such as mine being dismissed as unsophisticated proles. Even if that is not what you were trying to do... Matt Levine definitely comes off like that in his article.