By Gabe Minton

There is much change afoot in the mortgage industry—not the least of which is in the origination technology sector. Projects are having scope reductions, being deferred or cancelled altogether. From a loan product perspective, the industry is shifting from independent and customized to "canned" and predictable.

Software development has always been expensive; the more customized it is, the more expensive it is. When I lecture software-engineering classes, as an academic exercise I compare software engineering with civil engineering. Both are engineering disciplines with rigor and processes associated with the creation or engineering of the product they focus on (application or bridge), but software cannot be seen or touched. If a government or township undertakes construction of a bridge, it will very likely not cancel the project or reduce its scope if there are cost or schedule overruns. It certainly won't terminate the project before completion, and run cars and people over unfinished girders to get across. Yet this happens frequently in software development projects.

Project scope frequently is reduced (features are removed) and software is shipped early before completion or full user-acceptance testing by the customer. Why? I argue that it is because you cannot see or touch it. There is no way to see that only the girders are laid out for the bridge, that the architecture is sound and you can walk across, but more budget is needed to finish and allow people to cross safely. But I digress.

We have gone back in time in the mortgage industry. During the boom years (2002–2005), lots of capital was available; there were thousands of different loan products, and private-label securitizations were on the climb. This facilitated new levels of mortgage product independence by mortgage bankers and investors. Organizations desired flexibility and customization options in their origination and processing technology solutions. Niche players formed new organizations to manage the sheer number of products that were available (e.g., stand-alone product and pricing engine services).

Overall, a multitude of opportunities for new companies and new ideas were created—many of them good. New business models emerged, including transactional "pay to play" models. As long as the volumes were high, companies thrived. Now the trend is reversed. Organizations still want rich technology, but there is more emphasis on fewer products, Web connectivity, and strong regulatory and investor compliance. A typical lending organization now offers far fewer products, with emphasis on Federal Housing Administration (FHA)/Department of Veterans Affairs (VA) and government-sponsored enterprise (GSE) products. The private-label secondary market that invested in the variety of loan products available is gone, for the most part.

Originators want solutions that offer a lot of rigor with additional checks and balances, but they also want it "off the shelf." They don't have the budget or time to heavily configure their own solution (reverse-mortgage lending is a slightly different story). A lot of folks are reacquainting themselves with FHA lending (do you have a CHUMS [computerized homes underwriting management system] number from yesteryear?). There are still projects to replace antiquated or ailing systems, or systems that were provided by companies (especially niche players) that have phased out product lines or are no longer in business.

Given this, there are still technologies that are very important—but for different reasons. Imaging and eMortgages continue to rise, so new technology is still needed to produce and consume these types of files. Dynamic rules-based workflow is still very popular in origination systems, but the workflow is now more constrained with more checks and balances on fewer products instead of focusing on handling hundreds (or thousands) of flexible product and pricing definitions.

There is a renewed focus on the retail channel and away from the wholesale and correspondent channels. The move to Internet-based computing is still in force; it is just aimed at a mobile lender work force instead of enabling tens of thousands of independent brokers.

There is still an overall architectural "update" happening in the mortgage industry (and other verticals). That update includes companies that are moving off of Microsoft Windows-specific clientserver applications to Web-based applications. This includes all the buzzwords you have been reading about: XML (extensible markup language), Web Services, .Net", Enterprise Java", lightweight Web-based client applications and more.

Security investment is on the rise—not just because of increased legislative and regulatory requirements (Gramm-Leach-Bliley Act, Sarbanes-Oxley, California Security Breach Information Act [SB-1386], etc.), but also because the more we move our applications and data to the Internet, the more additional security requirements need to be implemented.

The Internet was not designed to be inherently secure, so the security layer has been added at the hardware or application level. Fraud detection continues to be a hot button. Companies are augmenting their origination solutions with more workflow (e.g., additional red flag-raising and handling) in this area. Business users are keen to take advantage of the latest in Microsoft graphical user interface (GUI) trends (think of MS Office 2007's "ribbon control").

This is not to say that someday the wild world of different mortgage products won't come back (at least to some extent)—this is a cyclical business, after all. In fact, in the reverse-lending arena, one could argue that the lenders are at the precipice of inventing new products within the context of the whole reverse-mortgage life cycle. For now, demand—in terms of the type of origination technology solutions and what they are focused on—has shifted.

Where is new technology being developed? Look no further than all the activity and implementation in loan modification, servicing and default-management solutions.