The devil is really in the detail. The most important being that it doesn’t just cover securitisations, but any “credit risk transfer product”. This explicitly includes credit derivatives and syndication. The impact of the proposal goes much, much further than “complicated securitsations”.

The second important thing to consider is that it doesn’t just require the originator/arranger/servicer (as relevant) to hold 10% of the total exposure. It requires the originator/arranger/servicer to hold 10% of each tranche it wants to sell to a bank.

The third is that there is nothing in here to stop banks from buying subprime debt per se. If a bank with a large balance sheet is active in the non-conforming sector (like HBOS, say) they could probably to retain the required 10%. Of course, big banks like that would probably prefer to fund those mortgages on balance sheet anyway, given the spreads on non-conforming RMBS. And specialist lenders without big balance sheets behind them will find things very difficult. This applies to other, prime sectors as well – conduit CMBS being the most obvious example. This market is dead for the foreseeable future anyway, given the condition of banks’ balance sheets, but the proposal almost guarantees that it won’t come back.

Finally, this rule would have made almost no difference to the IKB situation – IKB held almost all of its subprime assets through an off-balance conduit and an SIV, which wouldn’t have fallen under the provisions of the directive.

This makes it an effective rule regardless of what non-European governments decide to do about the securitization market, even if they decide not to act. The result would be that European Banks could no longer buy nonconforming securities from anywhere in the world, leading to some protection of European Banks from such toxic waste as repackaged US subprime debt or whatever the future will bring.

In fact, the article emphasizes the bailout of banks like IKB at (German) taxpayers’ expense because of their extensive investments in US subprime debt.

A second FTD article calls the proposal clever, which seems to be a reasonable assessment.

In sum: (European) Banks would probably be less risky if they could only buy complicated securitizations where the selling bank kept 10% on the books.

Now some deals were structured so that the packager kept the equity, but that was more than a bit of optics. They took so much in fees off the top that the equity could be and turned out to have been worth zero and they still did very well. At least until their trading inventory plummeted.