Delinquency Rates: Basics

By Cameron Polo

The role individual homeowners played in contributing to the Global Financial Crisis of 2007-09 is often measured by the number of defaults and foreclosures. While many Americans did lose their home – and these stories garnered significant media coverage – a more accurate assessment of how homeowners influenced the crisis can be found by looking at delinquency data, meaning the percentage of Americans who were late on their mortgage payment at any given time. Not every delinquency resulted in a foreclosure, but every delinquency both endangered the borrower, and represented a cash flow forgone, impacting mortgage servicers and the ultimate holders of mortgage assets.

To better understand how the delinquency rate in North Carolina evolved in the years leading up to the crisis, the American Predatory Lending team reviewed the National Delinquency Survey, generously provided to us by the Mortgage Bankers Association (MBA). Specifically, we reviewed a quarterly survey of mortgage loans in North Carolina. The graph below highlights the rapid increase in the number of loans showing up in the delinquency survey leading up to the crisis.


In just seven years, the number of loans surveyed  jumped from 668,000 to 1,415,000 – an increase of over 200 percent, capturing the drastic rise in originations as the Mortgage Bankers Association retained consistent sampling criteria. The increase in mortgage loans mirrors the increase in house prices, both of which were aided by the predatory lending practices documented in our oral history interviews. Many of these predatory practices were typically geared towards borrowers who may not otherwise have qualified for a mortgage loan, a key reason that journalists, politicians, and other observers often referred to the events of 2007-09 as the “subprime mortgage crisis.” However, this label is inaccurate.


As the above graph demonstrates, subprime made up a very small part of the overall mortgage market. Even at its peak in 2007, prime loans outnumbered subprime by a factor of ten. The data makes clear that subprime alone could not have caused the crisis.

The crisis revealed that prime loans were not as safe as most bankers and regulators assumed. A key development involved the evolving definition of “prime” leading up to the crisis. Recognizing the enormous profit potential tied to the booming U.S. housing market, Wall Street firms began to purchase large amounts of subprime loans to package into mortgage backed securities (MBS) and Collateralized Debt Obligations (CDO). Wary of losing market share to what became known as the “private-label” market, the government sponsored enterprises (GSEs) Fannie Mae and Freddie Mac, began to loosen their restrictions on loans that they would guarantee, or purchase to then be packaged into their own “agency” MBS. Therefore, as the graph below demonstrates, the conforming loan limit – the amount above which the GSEs are forbidden by law from purchasing – rose steadily during the housing boom.  This amount plateaued at $417,000 after the last raise in 2006— following the crisis, the Housing and Economic Recovery Act of 2008 declared this level not to rise until average home prices had risen to pre-crisis levels.[1]


Banks and nonbank mortgage lenders now had an incentive to originate as many loans as possible because they knew they could easily sell them to Wall Street or the GSEs and pocket the fees. This “originate-to-distribute” model pushed lenders to make mortgage loans that they would otherwise have refused to make if they had to keep the loan on their own balance sheet. Coupled with the continued investment grade ratings assigned by rating agencies to MBS and CDOs, the expansion of “originate to distribute” loans steadily eroded underwriting standards throughout the country and pushed home prices to unsustainable levels. When the bubble burst, delinquencies spiked dramatically.

*This is part of a series. Click here to read the next article.*

[1] Federal Housing Finance Agency. “Conforming Loan Limits.” (accessed April 10, 2020).