This post is Part 2 of a series of 3. Part 1 gives an overview of the problems in the mortgage market, casting them in terms I think technology can go a long way toward helping to solve. Part 2 explains how and why the change I suggest — switching form a market in mortgage-backed securities to one in actual mortgages — would actually work. Part 3 addresses the security and privacy concerns associated with such a transition.
I’m arguing that the market in MBS (Mortgage-Backed Securities) should be replaced by a direct market in individual mortgages. From a data management standpoint, this is technologically straightforward:
- All relevant data already exists in transactional databases.
- While there are many databases, schemas are simple and, as such, similar.
- Update rates are fairly low — a few per month per mortgage, tops, and often only one as the payment comes in.
Probably the biggest complication — other than the knotty issues of privacy and security — is that when the data is aggregated and presented for analysis, the resulting databases will be quite large. Speaking purely technologically, even that’s not a big deal. (But it does mean that third-party information service providers will be needed to serve investors, and that opens a can of worms about issues of who will profit from the information and how. More about that subject in a future post.)
Note: The total number of mortgages in the world is in the hundreds of millions. Why? Well, how many people live in economies where mortgages are common? Probably well under 2 billion. How many of those live in households that own their homes? Probably in the range of a billion, or less. How many people are there on average in such households? By now you see where I’m going with this. And commercial mortgages are surely much smaller in number than residential.
What’s more, there’s a strong sanity check on the intuition — models to judge individual mortgages already exist. Specifically, they’re used to make lending decisions by banks. And you know what? Those models work a lot better than MBS investors’ models do.
Why do I say that? Well, the banks that made mortgage loans and just held onto them aren’t the ones that are in the news in this crisis. This strongly supports the theory that the mortgage crisis is based on mortgage originators making under-priced loans and then selling them off to “greater fools” in the form of mortgage-backed securities.
“Greater Fool Theory” investing isn’t bad for all the individuals who engage in it. But somebody winds up as the Greatest Fool, and that person or organization is extremely unhappy. One analogy would be that it’s a giant game of musical chairs, and the music stopped.
So the goal would be to start with the kinds of credit models the banks use, refine them on available data, and invest on their basis. Prices for mortgages would still fluctuate over time, but on the whole it would be much more accurate. Historical data would be needed to prime the models — analysis just of new mortgages would be way too slow — but of course that’s available too.
Major changes to today’s MBS market would naturally ensue. Complications include:
- Somebody has to hold those securities other than the financial beneficiaries. Those “somebodies” need to be regulated.
- Investors couldn’t rely on bond ratings any more to convince themselves they were being prudent. This is a GOOD change, considering the disasters that overly optimistic ratings helped cause, but it’s still a big one.
- Some kinds of mutual funds or pooled investment vehicles would have to be created to serve smaller investors.
That last one is actually a benefit. There could be region-specific funds (great for local loyalists). There could be funds specializing in particular income levels or home sizes. This would all do more good than harm. But regulators would definitely have some work to do.
All of this is straightforwardly doable, from both a technological and a legal/regulatory/commercial standpoint. It’s a big change — but big change is needed. It applies to new mortgages much more than existing ones — but new home sales are a huge part of what is needed to heal the economy. With trillions of dollars gone, and trillions more in lost GDP adding up, the effort and cash costs are worth it many times over.
But there’s one kind of objection such hand-waving cannot legitimately dismiss — privacy and security. So I’ll turn to those in the third post of this series.
Bottom line: The individual mortgage aftermarket idea is practical.
