As the paintball game speeds up and all major players dodge faster to avoid designation as bearers of systemic risk, let's not get fixated on "too big to fail" as the past or future cause of the disaster.

There is something worse than "too big to fail," namely, "too few to compete." The rating agencies, for instance, were not too big to fail — but they were too few to compete. The same is true for mortgage servicing.

A couple of years ago, a bank I was advising became disgusted with the in-house servicing that came to them with an acquisition. So, they put this servicing up for bid. There was only one bid, take it or leave it, from one of the biggies. They accepted this bid — to their subsequent regret.

Servicing cannot be taken away from the largest servicers. There is no one else with capacity to handle it, so these few are beyond competition and above the law. They don't have to be responsive to their customers, the loan owners.

The Treasury Department raises doubts about controls and performance at large servicers, and Sheila Bair and others seek to impose national servicing standards — sounding as if there had never been any! False. "Servicing standards" were defined and allegedly enforced by the usual too-few-to-compete suspects. Fannie Mae and Freddie Mac had whole volumes full of servicing standards. One hand washed the other. Furthermore, the "best and brightest" hadn't been fighting to get mortgage servicing jobs. This wasn't the most rewarding place to work.

How did we come to this — and how do we get out of it?

Before the "crisis," servicing was merely a cost center generating fixed fees, and only a small part of its activity related to default loans. Servicing primarily entailed sending out tens of millions of statements, accepting payments and providing customer service. Major opportunities to generate value through consumer marketing were squandered because of the resulting cost-minimization mentality. There were economies of scale, so servicing became highly concentrated.

Given this perspective, collection processes, skills and efficiencies that had emerged as innovations in younger, more default-intensive and higher volume portfolios such as credit cards — which rely primarily on highly motivated in-house servicing by the lender — reached mortgage servicing only very slowly.

In portfolios other than mortgage, serious attention is actually given to detecting and repressing fraud! But as Ben Bernanke and others point out, "we are not even able to reach many of these borrowers to talk to them about their mortgages." The implications of this fact for the prevalence of fraud, the quality of the loans and the servicing — and appropriate mortgage resolution — are systematically ignored.

In the crisis, servicing of defaulted mortgages rapidly grew, to become a systemically critical activity. But, existing servicing contracts don't align the interests of servicer and owner in default situations. These contracts also permit or even encourage fundamental and very costly servicer conflicts of interest. If the servicer is affiliated with the originator, then we can't expect the servicer to help the owners of the loans detect and respond to origination fraud.

The owners constitute a diffuse and ill-informed constituency, and the trustees who are supposed to represent them are cost-minimizers who see themselves as having neither an upside nor a downside — another fundamental structural defect. Hence, demand for effective and results-oriented default servicing is muffled.

What is tragic and avoidable here is that the government-sponsored entities that had the most to lose on behalf of the taxpayers have not set the priority and paid the price to develop and require better, cleaner, more verifiable default servicing. Here's how this might be fixed.

Standardize data formats to increase shiftability of servicing. Starting with the prevalent conforming categories, the relevant government authority can then begin measuring default servicing performance on a consistent basis, shifting servicing where needed to facilitate clean comparisons. Then publicly report the results.

Use this data to promote competition, to get servicing moved to where it will be performed better — and attract needed capital. Use the data also to generate new best practices, particularly as to contact strategy, detection of origination fraud and the logic driving choice of action on defaults. I've never heard anyone declare publicly that keeping people in their homes won't always produce the best results for mortgage owners. So I'll say it here.

Rely on the regulatory whip hand to enforce compliance with best practices. Eliminate the most egregious conflicts of interest. Separate servicing from origination for loans owned by third parties, and develop model servicing contracts that discourage servicers from optimizing their bottom line at the expense of the mortgage owners. Make trustees liable for their negligence.

Many people feel that in the auto industry crisis, government intervention was effective and valuable as a last resort. Perhaps something similar is needed in mortgage servicing, because competition has failed.

Andrew Kahr is a principal in Credit Builders, a financial product testing and development company. He was the founding chief executive of Providian Corp. and can be reached at

Subscribe Now

Access to a full range of industry content, analysis and expert commentary.

30-Day Free Trial

No credit card required. Access coverage of the securitization marketplace, including breaking news updated throughout the day.