Managing the Mortgage Life Cycle

For many generations, homeownership has been the primary means of building wealth in the U.S. It has also increased liquidity in the economy as households use home equity to finance major life events such as college education and retirement, and homebuilding sectors. The mortgage loan, the principal vehicle making all this possible, plays a more significant role in our financial system.

As a financial instrument, the mortgage is an underlying contract used to generate traded securities and insurance products, an asset that supports balance sheets of financial institutions and something that is used in the pricing of investment products and real assets. As such, a mortgage, over its life cycle, can be viewed and managed as a container of vital information on the dynamics of household spending and wealth accumulation. In its amended form, as the asset base is leveraged and securitized, with new data derived over the lifetime of the contract, the information is a key component of economic indicators.

Information surrounding mortgages is continually updated, amended and consumed in various analyses that have significant impact on liquidity and capital allocation in the economy. Understanding this information is crucial to managing aggregate risk for financial institutions. Unfortunately, that lesson came too late.

Recommended For You

Both the subprime crisis of 2006-2007 and the crisis of 2008 showed the significance of mortgage debt in the functioning of the U.S. financial system. The total outstanding mortgage debt in Q1 2016 was $13.85 trillion, with federal agencies holding $5 trillion, depository institutions $4.8 trillion, insurance companies $430 billion, and pools and trusts $2.7 trillion (according to the Federal Reserve Board of Governors). Almost 80% of total household debt is mortgage loans. Of the 8.5 trillion mortgage-backed securities, seven trillion are held by government agencies. Household mortgage debt doubled between 2000 and 2006, although since 2012, it has increased by only 1% while interest rates since 2006 have dropped from an average of 7.6% to 3.85%. These are significant statistics against a backdrop of a U.S. GDP of $18.5 trillion.

The subsequent post mortem analysis on the financial crisis exposed a number of deficiencies across the system. Namely, an inability to manage that information flow, and to predict default rates as household income and property valuations changed dramatically, as balance sheet risks of institutions rose rapidly. The valuations and pricing of mortgage-backed securities and the correlations to underlying asset prices and volatility were frequently off the mark. And, it was not unusual to have a complete view of credit derivative contracts with all changes to terms and conditions, and to the counterparty obligations and exposures.

From the approval of a mortgage application to payoff, a history of contractual events – such as refinancing or secondary loans – end up generating a lot of new data that becomes available to insurance companies, investment firms and financial institutions. However, much of that data comes from manual tasks on paper – and is stored in many channels.

The decisions on portfolio strategy, collateral management, monitoring of default risks and investment products through securitization, pricing and more rely on this data flow and incorporation of the acquired insight with the general economic indicators. The cost of not keeping the complete relevant history and significant correlations of mortgage data in full view was the subprime crisis.

The systems and technology infrastructure that enable capture and consolidation of data in an accessible, consumable store takes time. Even now, Fannie Mae and Freddie Mac are not yet in a position to monitor their own holdings as tightly as they would like and continue to explore alternatives. However, a shorter-term approach to enhance the integrity of the system would be to manage metadata and improve access to distribution networks so that an on-demand, scenario-based analysis can be accomplished by most institutions.

If we are to imagine an environment where information is more fluid and accessible, it would be possible to more credibly answer questions on topics such as:

Default probabilities by region and asset price level, and balance sheet exposure;
Default probabilities of the counterparties;
Collateral valuations and implications on asset liability management;
Liquidity risks and liquidity shortfalls;
Sensitivity of household income fluctuations and unemployment rates on asset-backed security pricing; and
Accuracy of pricing and valuation models, and assessment of modeling risks.

A consolidated repository for mortgage data and household data by the participants in the mortgage market would be a solid first step to remove the operational (and reputational) risks associated with incomplete or out-of-date information. The analytical models that support processes tracking loan payments and default risks would be improved with updated household data, employment statistics and local asset price fluctuations. The trading and portfolio allocations for mortgage-backed securities (whether they are derived on principal or payments) would rely on better and more timely profiling of loan valuations.

The exchange of data across institutions and consolidated feeds into the regulatory institutions would also allow for better supervision – from initiation to servicing and securitization – of both existing and new loans.

The management and supervision of $13 trillion worth of the nation's asset base will be more effective and transparent depending on the speed that the industry moves away from siloes and fragmentation of related data and events. The health of the economy requires it.