Lunchtime Data Talk. Housing Finance Policy Center. Mortgage Origination Pricing and Volume: More than You Ever Wanted to Know

Transcription

1 Housing Finance Policy Center Lunchtime Data Talk Mortgage Origination Pricing and Volume: More than You Ever Wanted to Know Frank Nothaft, Freddie Mac Mike Fratantoni, Mortgage Bankers Association October 16,

2 Primary Mortgage Market Survey and Refinancing Report Urban Institute Lunchtime Data Seminar October 16, 2013 Frank E. Nothaft Chief Economist 2

3 Data, Data, Data Primary Mortgage Market Survey Longest weekly national survey of rates: 42½ years and counting 30-year & 15-year FRMs, 5/1 & 1-year Treasury-indexed ARMs Refinance Report Cash-out, Cash-in, Rate & Term Product transition: FRM vs ARM, 30-year vs shorter-term Single-family Mortgage Originations: Science and Art Benchmark to latest HMDA Potpourri of data to estimate current-year volume Office of the Chief Economist 3

4 Primary Mortgage Market Survey (PMMS) Freddie Mac introduced the weekly PMMS in April 1971 Business need was real-time market rates to set secondary market pricing Today the PMMS surveys over 100 lenders (representative mix) each week to obtain primary market quotes on 4 conventional products 30-year FRM, 15-year FRM, 5/1 Treasury-indexed Hybrid ARM, 1-year Treasury-indexed ARM 80% LTV, prime-credit, purchase-money, conforming ($200,000) Survey conducted Monday-Wednesday; Results released every Thursday (day before if holiday) Other PMMS products include monthly Refinance Share and ARM share of applications and annual ARM pricing survey Office of the Chief Economist 4

5 PMMS: Longest Weekly, National Survey of Rates Percent 20.0 Percent year Fixed-Rate Mortgage year ARM Source: Freddie Mac, Primary Mortgage Market Survey Office of the Chief Economist 5

6 PMMS Is Widely Used and Cited The Conventional Mortgage Rate on Fed s H.15 Ingredient for defining high-priced loans FFIEC s Average Prime Offer Rate (APOR) used for determining high-priced loans for HMDA reporting and HOEPA statements ( New York statute uses PMMS % to define subprime home loans (Banking Law 6-m) Widely cited in media Nearly 1,000 mentions every week and many, many more tweets Office of the Chief Economist 6

7 Selected Contract Rates on Conforming 30-Year Fixed Mortgages Percent 6.0 PMMS MIRS (t-1) MBA HSH Jul-13 Apr-13 Jan-13 Oct-12 Jul-12 Apr-12 Jan-12 Oct-11 Jul-11 Apr-11 Jan-11 Oct-10 Jul-10 Apr-10 Jan-10 Oct-09 Jul-09 Apr-09 Jan-09 Sources: Freddie Mac, Federal Housing Finance Agency, Mortgage Bakers Association and HSH Associates. Office of the Chief Economist 7

8 Selected Fees and Points on Conforming 30-Year Fixed Mortgages Percent PMMS MIRS(t-1) MBA HSH Jul-13 Apr-13 Jan-13 Oct-12 Jul-12 Apr-12 Jan-12 Oct-11 Jul-11 Apr-11 Jan-11 Oct-10 Jul-10 Apr-10 Jan-10 Oct-09 Jul-09 Apr-09 Jan-09 Sources: Freddie Mac, Federal Housing Finance Agency, Mortgage Bakers Association and HSH Associates. Office of the Chief Economist 8

9 HSH Conforming 30-year Fixed Mortgage Contract Rate Spread to PMMS Basis Points (HSH less PMMS) Average spread 12 basis points Sources: Freddie Mac and HSH Associates. Office of the Chief Economist 9

10 HSH Conforming 30-year Fixed Mortgage Effective Contract Spread to PMMS Basis Points (HSH less PMMS, adjusted for points) Average spread 0.6 basis points Sources: Freddie Mac and HSH Associates; assumes the MBA s conversion factor between The contract rate and fees & points Office of the Chief Economist 10

11 Conversion Factor Between Fees & Points and Interest Rate for Conforming 30-Year FRMs Conversion Factor ( DVO1 ) 9 8 MIRS MBA 3 Jul-13 Apr-13 Jan-13 Oct-12 Jul-12 Apr-12 Jan-12 Oct-11 Jul-11 Apr-11 Jan-11 Oct-10 Jul-10 Apr-10 Jan-10 Oct-09 Jul-09 Apr-09 Jan-09 Sources: Mortgage Bankers Association, FHFA Office of the Chief Economist 11

12 Why are PMMS Contract Rates Lower Than MBA s? Office of the Chief Economist 12

13 Loan Product, Purpose, Amount, and Credit Characteristics Explain Most of Difference Office of the Chief Economist 13

14 Need to Adjust for Points as Well Office of the Chief Economist 14

15 Adjusting for Points, Mean Difference = 5 bps Office of the Chief Economist 15

16 Quarterly Refinance Report Built on repeat transactions loan data Freddie Mac owns two successive first liens on house Second transaction is a refinance Compare the paid-off loan with the new loan Pre- and post-loan balance ( Cash-out, Cash-in, Rate & Term ) Pre- and post-contract rate Pre- and post-loan products ( product transition ) Estimates of aggregate home equity extraction Located at: Office of the Chief Economist 16

17 New Mortgage Rate Lows Create Refinance Booms Refinance Share of Applications (Percent) Boom Boom Boom 30-Year FRM Rate (Percent) Boom Refinance Share of Applications (left axis) FRM Rate (right axis) 4 3 Sources: Freddie Mac s Primary Mortgage Market Survey. Office of the Chief Economist 17

18 % Shortened Loan Term When Refinancing 80% 70% 60% 65% 50% 40% 30% 31% 20% 10% 0% 4% Shorter Term Same Term Longer Term Source: Freddie Mac Product Transition data. Balloons are excluded from the analysis. Percentages rounded to nearest whole number. Office of the Chief Economist 18

19 % Took Cash-Out At Refinance vs. 89% in Q % 90% 80% 83% 70% 60% 50% 40% 30% 20% 10% 0% 15% 2% Cash-Out Share No Change Cash-In Share Source: Freddie Mac Cash-Out Refinance Report data. Second quarter Percentages rounded to the nearest whole number. Office of the Chief Economist 19

20 Net Equity Cashed-Out in Refinance of Prime Conventional Loans (billions, In 2012 dollars) Borrowers Cashed-Out About $9 Billion When Refinancing in 2013Q2 $120 Dollars of Equity Cashed-Out (Left) 90 $100 $80 $60 $40 $20 Refinance Share of Applications (Right) Refinance Share of Mortgage Applications (In Percent) $0 0 Source: Freddie Mac, Bureau of Labor Statistics (Consumer Price Index) Second Quarter 2013, Equity cashed-out adjusted for inflation. Office of the Chief Economist 20

21 Discussion of MBA Mortgage Market Data October 16, 2013 Mike Fratantoni Vice President, Single-family Research & Policy Development Presented by David H. Stevens President, Mortgage Bankers Association 21

22 Overview For more than twenty years, MBA s Weekly Applications Survey (or the Survey, or Weekly Apps) has provided a timely indicator of housing and mortgage market activity. Since the survey s inception in 1990, its indices have been a leading indicator of housing and mortgage finance activity. All unadjusted indexes are set equal to for the week of March 16, Beginning with data from the week of September 16, 2011, released on September 21, 2011, MBA s Weekly Applications Survey was enhanced along a number of dimensions: Due to recruitment of new participants, MBA estimates the survey now captures more than 75% of all retail and direct channel mortgage applications, compared to 50% previously. There is now significantly more detail regarding the composition of applications, including the geographic and product mix of applications. MBA has collected additional information regarding mortgage rates. In addition to the rates previously reported, MBA now also reports on 5/1 ARM rates and 30-year fixed rates for jumbo loans. 22

23 Channel Mix: Large Lenders (% of Total Production $) Direct 8% 7% 9% 8% 18% 21% 18% 14% 14% 13% 13% 17% 15% 23% 25% Corr 36% 33% 33% 39% 31% 29% 37% 40% 31% 45% Broker 33% 24% 19% 11% 11% Retail 39% 39% 40% 39% 29% 22% 31% 28% 27% 33% Source: MBA/STRATMOR Peer Group Survey 23

24 2012 Channel Mix by Peer Group 2% 13% Consumer Direct 23% 10% 31% 17% Corr 10% 36% Broker 20% Retail 8% 77% 80% 32% 39% Large Banks Mid-Size Banks Large Independents Mid-Size Independents Source: MBA/STRATMOR Peer Group Survey 24

25 Purchase Index - Historical Source: Mortgage Bankers Association 25

26 Purchase Index Recent Activity Source: Mortgage Bankers Association 26

27 Purchase Apps and Home Sales Historically, the purchase applications index has been highly correlated with and has led home sales data by 4-6 weeks. In the past few years, a rising cash share of purchases, changes in the share of applications going through the retail and consumer direct channels, and changes in the home sales data have led to some separation. 27

28 Refinance Index - Historical Source: Mortgage Bankers Association 28

29 Refinance Index Recent Activity Source: Mortgage Bankers Association 29

30 Jan Apr Jul Oct Jan Apr Jul Oct Jan Apr Jul Oct Jan Apr Jul Oct Jan Apr Jul Oct Jan Apr Jul Oct Jan Apr Jul Oct Jan Apr Jul Oct Jan Apr Jul Oct Jan Apr Jul Oct Jan Apr Jul Oct Jan Apr Jul Prepay Speeds and Refinance Apps MBA Index Applications for Refinancing FNMA MBS Annualized Liquidation Rates Sources: Mortgage Bankers Association and Fannie Mae Monthly Volume Summaries 30

31 Additional Data Available on a Monthly Basis Composition report for US Monthly state profile of mortgage activity Source: Mortgage Bankers Association 31

32 January 2012 February 2012 March 2012 April 2012 May 2012 June 2012 July 2012 August 2012 September 2012 October 2012 November 2012 December 2012 January 2013 February 2013 March 2013 April 2013 May 2013 June 2013 July 2013 August 2013 September 2013 State-level Refinance Apps Trend 250% Year over Year Change in Refinance Applications Top 5 Refinance Volume States 200% 150% 100% 50% 0% -50% -100% CA FL TX NY GA Source: Mortgage Bankers Association 32

33 State-level Purchase Apps Trend 60% 50% 40% 30% 20% 10% Purchase Share versus 2012 Top 5 Purchase Volume States 0% CA FL TX NY WA September 2012 September 2013 Source: Mortgage Bankers Association 33

34 Discussion of New Mortgage Rate Series Each week mortgage lenders report their weekly interest rates and associated points in the Survey along with their mortgage application volumes. For conventional loans, the contract interest rates are for an 80% loan-to-value (LTV) loan. FHA loans have higher average LTVs. The points measure includes both origination fees and discount points. The following interest rates are covered: FRM 30-Year (Conforming) FRM 15-Year FRM 30-year Jumbo 5/1 ARM FHA 203(b) Survey participants must contribute a minimum number of each loan type to enter the calculation. 34

35 Comparison of MBA and Freddie Mac Survey 30-Year Fixed Rates Source: Mortgage Bankers Association, Freddie Mac 35

36 Jumbo-Conforming Spread 36

37 June 2009 August 2009 October 2009 December 2009 February 2010 April 2010 June 2010 August 2010 October 2010 December 2010 February 2011 April 2011 June 2011 August 2011 October 2011 December 2011 February 2012 April 2012 June 2012 August 2012 October 2012 December 2012 February 2013 April 2013 June 2013 August 2013 Jumbos Have a Lower Refi Share 100.0% Refi Share of Conforming Loans and Refi Share of Jumbo Loans 90.0% 80.0% 70.0% 60.0% 50.0% 40.0% 30.0% RefiShareofConfLoans RefiShareofJumboLoans 37

38 June 2009 July 2009 August 2009 September 2009 October 2009 November 2009 December 2009 January 2010 February 2010 March 2010 April 2010 May 2010 June 2010 July 2010 August 2010 September 2010 October 2010 November 2010 December 2010 January 2011 February 2011 March 2011 April 2011 May 2011 June 2011 July 2011 August 2011 September 2011 October 2011 November 2011 December 2011 January 2012 February 2012 March 2012 April 2012 May 2012 June 2012 July 2012 August 2012 September 2012 October 2012 November 2012 December 2012 January 2013 February 2013 March 2013 April 2013 May 2013 June 2013 July 2013 August 2013 ARM Share 35.0% ARM Share of Jumbo, Conforming and Total Loans 30.0% 25.0% 20.0% 15.0% 10.0% 5.0% 0.0% ARMShareofJumboLoans ARMShareofConfLoans ARMShareofTotalLoans 38

39 Credit Availability Index The Mortgage Credit Availability Index (MCAI) is calculated using several quantitative metrics related to borrower eligibility Credit score, loan type, loan-to-value ratio, etc. AllRegs Market Clarity Data MBA Proprietary Model Mortgage Credit Availability Index Metrics are combined to calculate the MCAI, a summary measure which indicates the availability of mortgage credit at a point in time. MCAI is a measure of mortgage credit supplied to the market. Unlike survey measures which only capture whether lending provisions are tightening or loosening, the MCAI provides a standardized quantitative metric that is solely focused on mortgage credit. 39

40 AllRegs Data, Processing by MBA Transformation of AllRegs Market Clarity Data by MBA analyst. TERM: Original = 5,10,15,30 Transformation: Min Term=5; Max Term=30 PURPOSE: Original = Cash Out Refinance, Purchase, Rate and Term Refinance Transformation: Allows Cash Out Granular data is categorized for standardization and analysis. MAX LTV Original = 95.6 OR 95 OR 92 OR 98.5, etc.. Transformation = LTV <=90 LTV >90 and <=95 LTV >95 40

41 7 Dimensions of Risk Weighting Amortization IO/Amortization Loan Purpose Product_Type FIXED ARM BALLOON Weight LTV / Occupancy Joint Int_Only_Flag Product_Type Weight NON-IO FIXED NON-IO ARM NON-IO BALLOON INTEREST ONLY FIXED INTEREST ONLY ARM INTEREST ONLY BALLOON Doc.Type Doc_Type_ROLLUP FULL/ALT DOC STATED DOCUMENTATION (Anywhere the word *Stated* appears) NO DOC (Anywhere the word *NO* is mentioned, excludes any entries with the word *STATED*) FICO / LTV Combo purpose_rollup Allows Rate/Term Refi, Not Cash Out Allows Cash Out Refi Purchase Transactions Only Loan Term term_rollup Max Term <= 30 Max Term >30 Weight Weight Weight Max_LTV_Bracket occupancy_rollup Weight LTV <=90 Allows Investor Homes LTV >90 and <=95 Allows Investor Homes LTV >95 Allows Investor Homes LTV <=90 Allows Second Homes, Not Investor LTV >90 and <=95 Allows Second Homes, Not Investor LTV >95 Allows Second Homes, Not Investor LTV <=90 Owner Occupied Only LTV >90 and <=95 Owner Occupied Only LTV >95 Owner Occupied Only Max_LTV_Bracket Min_FICO_Bracket Weight LTV >95 FICO <=620 LTV <=90 FICO <=620 LTV >90 and <=95 FICO <=620 LTV >95 FICO >620 and <=680 LTV >90 and <=95 FICO >620 and <=680 LTV <=90 FICO >620 and <=680 LTV <=90 FICO >680 LTV >95 FICO >680 LTV >90 and <=95 FICO >680 41

42 Historical Simulation 3/1/2007 5/1/2007 7/1/2007 9/1/ /1/2007 1/1/2008 3/1/2008 5/1/2008 7/1/2008 9/1/ /1/2008 1/1/2009 3/1/2009 5/1/2009 7/1/2009 9/1/ /1/2009 1/1/2010 3/1/2010 5/1/2010 7/1/2010 9/1/ /1/2010 1/1/2011 3/1/2011 5/1/2011 7/1/2011 9/1/ /1/2011 1/1/2012 3/1/2012 5/1/2012 7/1/2012 9/1/ /1/2012 1/1/2013 3/1/2013 5/1/2013 Histoical Simulation: Mortgage Credit Availability Index, Index Level by Month (NSA, 3/2012=100) Reading the Mortgage Credit Availability Index Lower Index = Less Credit Available Higher Index = More Credit Available Simulated Index Value Relative 3/2012=100 Base Period Current Index Levels (3/2012 = 100) Source: Mortgage Bankers Association, AllRegs Market Clarity AllRegs and MBA examined guidelines from the period and estimate that the Mortgage Credit Availability Index was probably somewhere between 750 and 850 during this period. This was due to the high availability of low-doc; no-doc; stated income; interest-only loans. Also, the lending environment during that time made drawing cash against the equity in a home a common and accessible industry trend, making credit more available during that time while also carrying a high amount of risk. 42

43 Current Trends in the MCAI The MCAI is benchmarked to an index value of 100 in March Relative to the base period (March 2012), index values above 100 represent greater availability of mortgage credit while index values below 100 are an indication that mortgage lending standards have tightened. Source: MBA MCAI 43

44 Builder Application Survey The MBA s Builder Application Survey (BAS) measures loan application activity received by lenders affiliated with or who work in cooperation with home builders for loans on new single-family properties. This survey focuses solely on the new home sales market. The monthly BAS Press Release will track changes in national application volume, average loan size and loan products used for new home sales purchases. The MBA will also release a New Home Sales estimate. Release Date Second Week of the Month As the survey continues to expand, the MBA expects to phase in more detailed information including state- and metropolitan-level statistics. 44

45 Comparison of Index Levels BAS Index Level Jan-11 Feb-11 Mar-11 Apr-11 May-11 Jun-11 Jul-11 Aug-11 Sep-11 Oct-11 Nov-11 Dec-11 Jan-12 Feb-12 Mar-12 Apr-12 May-12 Jun-12 Jul-12 Aug-12 Sep-12 Oct-12 Nov-12 Dec-12 Jan-13 Feb-13 Mar-13 Apr-13 May-13 Jun-13 Jul-13 Aug-13 New Home Market versus Total Market WAS Index Level Date Builder Application Survey (NSA) Weekly Application Survey Purchase Index (NSA) 45

46 New Home Sales Estimate Jan_11 Feb_11 Mar_11 Apr_11 May_11 June_11 Jul_11 Aug_11 Sep_11 Oct_11 Nov_11 Dec_11 Jan_12 Feb_12 Mar_12 Apr_12 May_12 June_12 Jul_12 Aug_12 Sep_12 Oct_12 Nov_12 Dec_12 Jan_13 Feb_13 Mar_13 Apr_13 May_13 Jun_13 Jul_13 Aug_13 50,000 Census New Home Sales vs MBA New Homes Sales Estimate Monthly, NSA 45,000 40,000 35,000 30,000 25,000 20,000 Census NHS (NSA) MBA NHS Prediction 46

47 Builder App Trends by State 47

48 Single-family Originations: Science and Art Benchmark to latest HMDA Compare volume reported sold to Freddie Mac with our purchase volume 2012 originations sold to Freddie Mac in originations sold to Freddie Mac in 2012 by a wholesaler HMDA volumes totaled about 95% of Freddie Mac s purchase volume Conventional originations reported in 2012 HMDA = $1.66 trillion, thus $1.66/0.95 = $1.75 trillion conventional originations (1 st and junior liens) Add FHA+VA+USDA = $0.38 trillion, puts 2012 single-family at $2.13 trillion MBA data helps gauge recent activity Dollar application volumes not released by MBA Compare conventional applications in 2012 HMDA with average MBA index for conventional applications Use ratio to convert current-year index into application volume Office of the Chief Economist 48

49 $4,000 $3,750 $3,500 $3,250 $3,000 $2,750 $2,500 $2,250 $2,000 $1,750 $1,500 $1,250 $1,000 $750 $500 $250 $0 Mortgage Origination Volume Is Large and Variable Total Single-Family Mortgage Originations (Billions of Dollars) FHA & VA Conventional Refi Coventional Purchase Forecast 25% Drop '98 '99 '00 '01 '02 '03 '04 '05 '06 '07 '08 '09 '10 '11 '12 '13 '14 Sources: Freddie Mac, HUD, VA 49

50 Weekly Conventional Application Volume 1 $ Billions Purchase up 12% Refi down 68% (same week, Y-O-Y) Conv Refi, $17.8 B /5/12 11/5/12 12/5/12 1/5/13 2/5/13 3/5/13 4/5/13 5/5/13 6/5/13 7/5/13 8/5/13 9/5/13 Conv Purch, $12.1 B Source: MBA Weekly Applications Survey, Conventional Market (NSA), HMDA, Freddie Mac 1 The MBA Survey does not release dollar application volume, but using 2012 HMDA as a base year, we estimate market level application dollar volume. 50

51 Where to Get More Information Look for regular updates to our economic forecast, commentary and data at Contact us at Opinions, estimates, forecasts and other views contained in this document are those of Freddie Mac's Office of the Chief Economist, do not necessarily represent the views of Freddie Mac or its management, should not be construed as indicating Freddie Mac's business prospects or expected results, and are subject to change without notice. Although the Office of the Chief Economist attempts to provide reliable, useful information, it does not guarantee that the information is accurate, current or suitable for any particular purpose. Information from this document may be used with proper attribution. Alteration of this document is prohibited by Freddie Mac. 51

52 Outlook for 2013 and GDP Growth 1.8% 1.7% 2.5% Inflation 1.9% 2.1% 1.9% Unemployment 8.1% 7.5% 7.0% Fed Funds 0.1% 0.2% 0.2% 10-year Treasury 1.8% 2.3% 3.0% 30-year Mortgage 3.7% 4.1% 4.9% Refi originations ($ B) 1, Purchase originations ($B) New Home sales (thousand) Existing home Sales (thousand) 4,634 5,070 5,362 Source: MBA August 2013 Forecast 52

53 Estimated Originations 1990 to 2014: SF Market Source: MBA August 2013 Mortgage Finance Forecast 53

54 Contact Information & MBA Resources Michael Fratantoni Vice President, Single-family Research and Policy Development (202) MBA Homepage: MBA Research: MBA Weekly Apps (Main page, FAQ, Methodology Document): Builder Application Survey: MCAI Product Page: 54

55 1