How Sound are the Fannie Mae and Freddie Mac Recoveries?

Fannie Mae and Freddie Mac (Fannie and Freddie) are the government-sponsored enterprises that back the majority of U.S. residential mortgage loans. In the last financial crisis, Fannie and Freddie played a large role by their purchase of inadequately qualified loans, subsequent bankruptcy, and bailout by the U.S. government. As a result, much has been written about the importance of monitoring the quality of mortgage loans purchased by Fannie and Freddie.

Recently, Fannie and Freddie have made new headlines due to its turnaround under government control. With the latest dividend payment, Fannie and Freddie have paid a total of $266 billion to the U.S. Treasury, making the bailout a tremendous success. In this article, the authors report our findings on the changes in Fannie and Freddie’s mortgage loan quality by analyzing over 15 million loans originated by Fannie and Freddie during the period of 2001-2015. Our results indicate an increase in the quality of post-crisis loans compared to pre-crisis loans, especially in debt-to-income ratio (DTI) and credit score (FICO) requirements. However, we also find that since 2008, their loan-to-value (LTV) requirements have been loosened, which has resulted in a steady increase of first-timer home-buyer loans.

The average characteristics of first-time home buyers are: higher LTV, lower FICO scores, and more likely to carry private mortgage insurance (PMI). As such, in the event of another downturn in the housing market, these loans are more susceptible to default. Accordingly, despite Fannie and Freddie’s recent positive performances, this article highlights the importance of monitoring the percentage of first-time homebuyer loans as well as LTV ratios within Fannie and Freddie’s mortgage portfolios.

Credit Supply of the Housing Market

The authors examined the credit supply of the housing market, which as of 2017 appears to have recovered from the 2008 financial crisis (see Case-Shiller Home Price Index in Figure 1). However, our question was whether the recovery was as complete as the Case-Shiller Home Price Index would appear to indicate, or whether there were still areas of vulnerability that would leave room for history to repeat itself.

Figure 1:

Case-Shiller 20-City Composite Home Price Index

Source: S&P Dow Jones Indices LLC

The majority of U.S. residential loans are sold in the secondary market where they are securitized, i.e., packaged into mortgage-backed securities and sold to investors. The two biggest securitizers are the government-sponsored enterprises Fannie Mae and Freddie Mac. As of 2010, 67 percent of residential mortgages were securitized, and 54 percent were securitized by Fannie and Freddie.[1] Fannie and Freddie were partially blamed for the 2008 crisis because, in alignment with the U.S. government’s first-time home ownership initiatives, Fannie and Freddie lent to less qualified borrowers. Hendershott, Hendershott and Shilling have provided a detailed summary of Fannie and Freddie’s role and influence in the housing market credit bubble.[2]

As the housing market has recovered, there have been stories about Fannie and Freddie’s tightening of their lending criteria.[3] But we wanted to know: how specifically have they changed their lending criteria after the financial crisis? Have loan qualifications indeed been made more stringent, and if so, in what ways?

The authors investigated these purportedly higher qualifications by looking at three major lending criteria: loan-to-value (LTV) ratio, debt-to-income (DTI) ratio, and credit score (FICO). These criteria are used to measure what have been considered the two driving factors behind a borrower’s choice to default, which are ability to pay and willingness to pay.

  • DTI and FICO score relate to the borrower’s ability to pay the payment on time. A considerable amount of research finds that a high DTI and a lower FICO score result in a higher probability of default (for example see references[4], [5]).
  • LTV measures the borrower’s willingness to pay. Where the house value falls below the loan balance, borrowers are likely to strategically default on mortgage payments, even if they are able to Even where house prices increase and the current LTV does not exactly reflect the initial LTV at the time of origination, empirical research shows that the initial LTV nonetheless has predictive power in default determination.[6]

The authors also examined how loan qualities were different for first-time home buyers (FTHB) compared to non-FTHBs. Historically, more than a quarter of all borrowers have been first-time home buyers. Research is inconclusive as to whether these FTHBs tend to default more.[7] However, statistically, FTHBs are likely to be younger households with less established financial resources. As a result, it may be argued that they have a higher risk of default (although this risk might be partially mitigated by the fact that they may also have greater ability for quicker re-employment).

Data & Analysis

Quality of Fannie Mae and Freddie Mac Loans

Fannie and Freddie have released loan-level origination and performance data from 2000 (Freddie Mac data is available from 1999; Fannie Mae data is available from 2000). Our study covered a total of 15 million loans, all of which are 30-year fixed rate purchase loans for primary residences. Figure 2 displays the total number of loan originations (left axis). The blue bars represent the number of FTHB loans, and the red bars are non-FTHB loans. The gray line on the right axis represents the FTHB ratio amongst the total number of loans. Figure 2 displays, as commonly known, the sharp decrease of loan originations following the financial crisis of 2008 until 2012. It also displays that as of 2015, loan origination activities have recovered to pre-crisis levels.

Notably, since 2006, more FTHBs have entered the housing market. The historical average of FTHBs is about 25 percent of total home buyers; in 2015, FTHBs were about 45 percent of the total number, which is a significant increase. Considering this noticeable change in the FTHB ratio, we investigated FTHBs separately in the subsequent loan quality analyses.

Figure 2:

Loan Originations

As a first step in our investigation, we examined the two ability to pay measures. Figure 3 shows the historical average DTI. The gray, blue and red lines indicate all FTHB and non-FTHB loans, respectively. From 2003, Fannie and Freddie allowed higher debt payment relative to income, which increased the risk level of a loan. By 2008, the year of the crisis, the average DTI was 39 percent. Following the market crash, Fannie and Freddie tightened the DTI to approximately 35 percent. However, more recent figures show that DTI is on the rise. Additionally, FTHBs appear to have lower DTI than non-FTHBs by approximately 2 percent. This gap is offset by the fact that FTHBs usually take out loans higher than 80 percent LTV, which then require private mortgage insurance (PMI), leading to additional monthly payments. For example, in 2015, more than 62 percent of FTHBs obtained loans with PMI. Annual PMI premiums range from 0.3 percent to 1.5 percent of the original loan amount.[8] Applying even a conservative premium of only 0.41 percent of annual PMI fee to the FTHB average loan size of $226,000, the resulting PMI premium would be $927 per year. Consequently, an average FTHB with an annual income of $41,000, would need to pay more than 2.3 percent of their income for the PMI; this would in turn offset the reported 2 percent lower DTI level. Thus, there would be almost no difference in DTI between FTHBs and non-FTHBs.

Figure 3:

Debt to Income (DTI)

Fannie and Freddie also significantly raised FICO standards. Before the financial crisis, the average FICO score of loan applicants was around 730. Currently, it hovers around 760. However, as Figure 4 shows, this standard appears to be dipping. Furthermore, FICO scores for FTHBs are more than 10 points lower than non-FTHBs.

Figure 4:

Credit Score (FICO)

Next, the authors looked at the willingness to pay measure, and Figure 5 shows the average LTV level. The LTV limit was at relatively low levels until 2011. Current LTV levels, however, are even higher than at the pre-crisis level. In 2015, the LTV level of FTHBs was 86 percent, which was 6 percent higher than non-FTHBs at 81 percent. This higher initial LTV level exposes FTHBs to the default incentives when house prices drop.

Figure 5:

Loan to Value (LTV)

If LTV is greater than 80 percent, lenders typically require PMI. This insurance premium as an additional expense reduces the borrower’s ability to pay. As FTHBs have higher LTV rates, more FTHBs have PMI payments. Figure 6 shows the ratio of loans that also had PMI. In 2015, almost 62 percent of FTHBs were paying for PMI, whereas only 45 percent of non-FTHBs had PMI. Thus, even if FTHBs had lower DTIs, this PMI added an extra payment obligation.

Figure 6:

Loans with Private Mortgage Insurance (PMI)

First Time Home Buyers (FTHB)

In this next section, the authors further investigated characteristics of FTHBs. Figure 7 shows borrowers’ average incomes, based on DTIs, loan sizes, interest rates and loan terms. FTHBs have lower average income than non-FTHBs: in 2015, these figures were $41,000 versus $44,000, respectively.

Figure 7:

Income

The relatively lower income of FTHBs indicated by the blue line in Figure 7 corresponds to the fact that FTHBs purchased lower-priced houses. Figure 8 displays the prices of these houses. As expected, because FTHBs were likely to be younger with less accumulated wealth and income, their purchase price level was lower than that of non-FTHBs. And, in 2015, the average house price of FTHBs was $260,000 while for non-FTHBs it was $310,000.

Figure 8:

House Price

House prices develop in different patterns depending on price tiers. See, for example, the Los Angeles Case-Shiller indexes for low-, middle- and high-tier price indexes in Figure 9. Low-tier prices are more volatile than the middle or high tier prices. Since FTHBs tend to buy lower price houses, FTHBs are more likely to go through unstable price changes in the future.

Figure 9:

Case-Shiller Index by Price Tier: Los Angeles

Source: S&P Dow Jones Indices LLC

Findings thus far suggest FTHBs represent a higher risk (higher LTV, lower FICO, and lower-tier housing market). This riskiness should be priced into their interest rates. Figure 10 displays the average interest rates of FTHBs and non-FTHBs. FTHBs were charged 0.05 percent higher interest rate on average, which merits further research as to whether this spread is sufficient to incorporate all risks associated with FTHBs.

Figure 10:

Interest Rate

 

Conclusion

Our investigations into Fanny and Freddie’s loans from 2001 to 2015 show that they did tighten standards for loan qualification after the financial crisis, as measured by higher average DTI and FICO scores. However, we also found that they did not tighten LTV requirements; in fact, LTV ratios have been rising since 2008. Furthermore, in the last few years, Fannie and Freddie have once again lowered their lending standards, likely due to criticisms that they have been too tight.[9] Thus, there is a growing risk attendant to the lowering of LTV standards, as it leaves Fannie and Freddie vulnerable for another housing market downturn.

The authors also found an increased percentage of first time home buyers (FTHBs). The characteristics of FTHBs are, generally, that they tend to borrow to their maximum capacity with a higher LTV, which requires extra PMI payments. Also, FTHBs tend to have lower FICO scores than non-FTHBs. Lastly, FTHBs tend to be in the low-tier housing market, which is more volatile than the middle or high-tier markets. Though on average FTHB loans have slightly higher interest rate (0.05 percent) than the non-FTHB loans, it is questionable whether this rate difference is sufficient to compensate for the higher risk associated with these borrowers.

Overall, despite recent positive performances by Fannie and Freddie, our study also revealed the growing need for increased monitoring of their deteriorating LTV standards and historically high number of FTHB loans. We also highlighted the need for further research as to whether the interest rate differential of 0.05 percent for FTHBs compared to non-FTHBs were sufficient to incorporate all risks associated with FTHBs.

 

[1] The Financial Crisis Inquiry Commission. (2011). “The Financial Crisis Inquiry Report”. January: p.69.

[2] Hendershott, P., Hendershott, R., & Shilling, J. (2010). The Mortgage Finance Bubble: Causes and Corrections. Journal of Housing Research 19: 1-16.

[3] MarketWatch. (2017). “Mortgage lending is so tight, more people aren’t even bothering to apply.” February. Accessed at http://www.marketwatch.com/story/mortgage-lending-is-so-tight-more-people-arent-even-bothering-to-apply-2017-02-14

[4] Holden, S., Kelly, A., McManus, D., Scharlemann, T., Singer, R., & Worth, J. D. (2012). The HAMP NPV Model: Development and Early Performance. Real Estate Economics, 40, S32-S64.

[5] Rose, M. J. (2012). Origination Channel, Prepayment Penalties and Default. Real Estate Economics, 40, 663-708.

[6] Deng, Y., Quigley, J. M., & Van Order, R. (2000). Mortgage Terminations, Heterogeneity and the Exercise of Mortgage Options. Econometrica, 68, 275–307.

[7] Ambrose, B. W., & Capone, C. A. (1998). Modeling the Conditional Probability of Foreclosure in the Context of Single-Family Mortgage Default Resolutions. Real Estate Economics, 26, 391-429.

[8] Lewis, H. (n.d.). “Private mortgage insurance, or PMI: Just the basics.” Accessed at http://www.bankrate.com/finance/mortgages/the-basics-of-private-mortgage-insurance-pmi.aspx

[9] MarketWatch, 2017.

2017 Volume 20 Issue 2

2017 Volume 20 Issue 1

2016 Volume 19 Issue 2

2016 Volume 19 Issue 1

2015 Volume 18 Issue 2

2015 Volume 18 Issue 1

2014 Volume 17 Issue 3

2014 Volume 17 Issue 2

2014 Volume 17 Issue 1

2013 Volume 16 Issue 3

2013 Volume 16 Issue 2

2013 Volume 16 Issue 1

2012 Volume 15 Issue 3

2012 Volume 15 Issue 2

2012 Volume 15 Issue 1

2011 Volume 14 Issue 4

2011 Volume 14 Issue 3

2011 Volume 14 Issue 2

2011 Volume 14 Issue 1

2010 Volume 13 Issue 4

2010 Volume 13 Issue 3

2010 Volume 13 Issue 2

2010 Volume 13 Issue 1

2009 Volume 12 Issue 4

2009 Volume 12 Issue 3

2009 Volume 12 Issue 2

2009 Volume 12 Issue 1

2008 Volume 11 Issue 4

2008 Volume 11 Issue 3

2008 Volume 11 Issue 2

2008 Volume 11 Issue 1

2007 Volume 10 Issue 4

2007 Volume 10 Issue 3

2007 Volume 10 Issue 2

2007 Volume 10 Issue 1

2006 Volume 9 Issue 4

2006 Volume 9 Issue 3

2006 Volume 9 Issue 2

2006 Volume 9 Issue 1

2005 Volume 8 Issue 4

2005 Volume 8 Issue 3

2005 Volume 8 Issue 2

2005 Volume 8 Issue 1

2004 Volume 7 Issue 3

2004 Volume 7 Issue 2

2004 Volume 7 Issue 1

2003 Volume 6 Issue 4

2003 Volume 6 Issue 3

2003 Volume 6 Issue 2

2003 Volume 6 Issue 1

2002 Volume 5 Issue 4

2002 Volume 5 Issue 3

2002 Volume 5 Issue 2

2002 Volume 5 Issue 1

2001 Volume 4 Issue 4

2001 Volume 4 Issue 3

2001 Volume 4 Issue 2

2001 Volume 4 Issue 1

2000 Volume 3 Issue 4

2000 Volume 3 Issue 3

2000 Volume 3 Issue 2

2000 Volume 3 Issue 1

1999 Volume 2 Issue 4

1999 Volume 2 Issue 3

1999 Volume 2 Issue 2

1999 Volume 2 Issue 1

1998 Volume 1 Issue 3

1998 Volume 1 Issue 2

1998 Volume 1 Issue 1