Is securitization the Betamax of finance?
What if securitization was just a convenient halfway house before the next evolution of finance?
Plenty of concepts in finance are old but, sometimes, technology allows them to happen. Securitization, in the sense of secured lending, has been around forever, but the process of combining thousands or hundreds of thousands of loans to back bonds needed processing power.
Figuring out the cashflows on securitizations simply could not happen before computers. Energetic clerking could perhaps track principal and interest payments, and perhaps even pass these through to the bonds.
But traders and investors could not feed in multiple factors like house prices, employment levels, wage growth, interest rates and the mortgage market into their cashflow models without computers, certainly not in real time, and with enough precision to value the deals and take a view.
This is loosely the tech-utopian view of finance — the pre-crash defence of securitization as a higher state, evolved method of financing appropriate for a digital age. Although the financial crisis exposed the hubris and blindness caused by such techno-optimism (when combined with greed and exploding leverage across the financial system) it’s not totally wrong.
Technology can and does enable new means of financing, and these are sometimes genuinely useful. So now, or soon, it will be time to abandon securitization.
Although technology allowed it to exist, it never completed the project.
In the early days of the private market, during the 1980s, hair and phones were big, but computing power was small, and cashflows were simple. Homogenised asset pools with simple waterfalls.
Chips got smaller, computer power kept doubling every 18 months, and master trusts weren’t far behind allowing much larger pools of revolving assets which can back discrete series of amortising bonds, offering flexible, programmatic issuance and allowing the sponsor to take back the risk of interest rate fluctuations.
New asset classes — and financial complexity — followed. Some of these, notoriously, blew up, but others offered neat solutions to real financial dilemmas.
Surely it’s a good thing if banks can now pay their employees in synthetic exposure to the mezzanine tranche of a bundle of fluctuating and revolving counterparty credit exposures (congrats, Credit Suisse)? But there needs to be some serious technological horsepower under the bonnet to value the instrument in real time, with many uncertainly correlated variables feeding into the instrument.
And yet the market still relies on large papery prospectuses, securities depositories, creaky settlement infrastructures, shell companies to contain assets, listing authorities, armies of lawyers, and rating agencies to rake over it all. The flow of payments through a securitization relies on a printed list of bullet points.
The fundamental purpose of securitization is converting illiquid loan pools into liquid, tradeable investments, but as technological systems become more capable of managing loan pools and displaying information in real time, there should be less and less need to actually bundle these into securities.
Some of the legal issues, like the ability to enforce security, perfect title to assets, or achieve ‘true sale’, are easiest to deal with using shell companies, but there’s nothing set in stone about the legal framework — plenty of countries have updated their securitization laws, over time, while there is a renewed push to look again at insolvency, tax and enforcement across Europe.
The marketplace lenders are showing the way. Many of them already have secondary markets for their loan exposures, but all of them have fast, flexible, natively electronic interfaces for making loans and assessing credit quality. Their loan books are already an open book to investors, unlike many of the banks with which they compete, some of which can scarcely identify what is on their own balance sheets.
Coupled with good modelling software, structuring securitization-style exposures from a whole loan book should become as simple as clicking and dragging. Assign the cashflows from this pool to that investor, under these conditions, and you’re done. Publish the loan data and the waterfalls and others in the market can value and bid for exposure to the pool.
With real time loan information, comprehensive loan level data and the processing power to manage this warehouse of information, why bother tranching, stamping with a rating, and going through the rigmarole of creating a listed security?
Of course, this would lose some of the magic of first generation securitization — namely, the creation of more bond-like, highly rated and highly standardised securities out of loan pools.
But this is exactly the part of the securitization which has been in the main regulatory firing line.
Regulators have never had a problem with investors buying exposure to loans that they understood; only with investors buying large amount of supposedly low-risk debt without taking the time to assess the underlying assets.
The enormous investments in loan monitoring, the increase in computing power and the march of the marketplace lenders all point in the same direction.
We’re not there yet, but it’s only a matter of time before the industry sheds its skin, and securitization evolves into its new incarnation.