Research. USA: Subprime mortgage market containment or contagion? March 30, Containment or contagion?

Transcription

1 Research March, 2 Peter Possing Andersen, , pa@danskebankdk USA: Subprime mortgage market containment or contagion? Problems in the US subprime mortgage market have fuelled fears of a broader contagion to the housing market and the economy in general We focus on three potential channels for a spillover from the subprime problems: a broad tightening of credit, rising foreclosure rates and the impact of Adjustable Rate Mortgage (ARM) resets Mortgage credit standards have been tightened in recent quarters And while we would not be surprised to see a further tightening of credit standards also outside the housing market during the coming quarters, the conditions for a broader credit meltdown are currently not present Rising foreclosure rates will put more homes up for sale This could potentially delay the stabilisation of the housing market However, the additional number of homes for sale in 2 will correspond to no more than one months turnover Hence, the overall impact on the housing market is not likely to be that great The bulk of the ARM resets will occur during 2 and 28 However, the negative impact will be limited, as it will correspond to less than 2% of aggregate personal income growth Moreover, the impact will hit less than % of US households Finally, it might be constructive to view the subprime problems as a symptom rather than a cause In the wake of 42bp of monetary tightening and a slowing housing market, it is understandable that some problems should show up in the riskiest section of the adjustable rate mortgage market While the problems in the US subprime mortgage market are a risk factor, we do not expect any deep macroeconomic ramifications via either a severe tightening of credit or via housing activity Containment is our call Containment or contagion? Subprime jitters cause concerns Troubles simmering in the US subprime mortgage market (see fact box page 2) have boiled over in recent months, fuelling fears of a broader contagion to the credit and housing markets and to the economy in general While there is presently a lot of focus on the housing market and housing data in general, this report deals only with the impact of the problems in the

2 RESEARCH MARCH, 2 subprime market Earlier reports covering US housing activity are available on our website HTwwwdanskebankcom/researchTH Below, we review the situation in the US subprime mortgage market and asses the macroeconomic consequences and risks connected with recent developments While we find that there will be some negative spillover from the problems in the subprime market, the effects are not likely to be disruptive for either the housing market or for the economy in general US subprime mortgage market To begin with, it could be useful to review a few facts about the US mortgage market This section covers most of the data already available in the HTUPresentation: The US subprime mortgage marketuth The table below illustrates the structure of the US mortgage market (note below table provides an explanation of the abbreviations) Table: Structure of the US mortgage market Loan types (Share of all serviced loans, %) ARM FRM Other Total Prime 1% % 2% 66% Sub-prime 66% 4% 22% 1% Other (VA, FHA) % Total 2% 84% 4% % - ex VA, FHA 22% 64% 126% Source: MBA (Mortgage Bankers Association, Ecowin) data for Q4 26 Note: ARM: Conventional adjustable rate mortgages, FRM: Conventional fixed rate mortgages, VA: Veterans Administration mortgages, FHA: Federal Housing Administration mortgages Both prime and subprime loans are facilitated by private agencies As the average size of individual loans differs, the table can only be used as a rough guide to the distribution of the dollar-value of all loan types from the outstanding total of USD billion ARM loans are typically slightly larger in dollar terms than FRM loans The dollar share of ARM loans is about 2% The US subprime mortgage market services households with a low credit quality (see box below) This section of the mortgage market has been growing in recent years, with market share increasing from 4% at the beginning of 2 to 1% in Q4 26 Sharply rising numbers of subprime mortgages is most likely a consequence of the very easy credit standards and the boom in house prices during this expansion More fundamentally though, the developments are attributable to the very easy monetary policy in the period 22- From this perspective, it seems only natural that some distress should show up when policy is tightened once more, as the subprime mortgage sector services only low quality borrowers Subprime mortgage market has grown 1 % % Subprime share of total mortgage loans Source: MBA & Ecowin 1 1 Within the subprime market, approximately 48% of the loans are conventional Adjustable Rate Mortgages (ARM), 6% are conventional Fixed Rate Mortgages (FRM), while 14% are other loan types (see table above) Fact box: The subprime mortgage market The subprime mortgage market is that section of the mortgage market devoted to borrowers with a lessthan-perfect or indeed a bad credit record It accounts for 1% of total mortgage market loans Subprime borrowers face higher costs due to their lower credit quality and higher risk of delinquency The subprime market provides the same range of loan types as the prime market, ie adjustable rate mortgages (ARM), fixed rates mortgages (FRM), hybrid ARMs, etc Recent data from the Mortgage Bankers Association indicate that problems in the subprime market are almost entirely related to the ARM segment This segment has seen a significant rise in both delinquency and foreclosure rates (see chart below) Rising subprime delinquencies 1 % % Subprime delinquencies 1 adjustable rate (ARM) 1 1 total fixed rate (FRM) In contrast, fixed rate mortgage (FRM) loans in the subprime market have seen almost no change in delinquency and foreclosure rates DANSKE BANK 2

3 RESEARCH MARCH, 2 2,2 1, 1,2, and rising subprime foreclosures,2 % Subprime foreclosures started % 2, adjustable rate (ARM) total fixed rate (FRM) ,2 2, 2,2 1, 1,2, Interestingly, the data pattern in the prime mortgage market largely mirrors this picture Rising delinquencies and foreclosures are entirely ARM related, while there has been no significant upward movement in the FRM figures (see charts below),,2 2, 2,2 1, ARM-related delinquencies 4,2 % Prime delinquencies % total fixed rate (FRM) adjustable rate (ARM) ,2,,2 2, 2,2 1, and foreclosures are rising in the prime market, % Prime foreclosures started %,4,4 adjustable rate (ARM),,,2,2,1, total fixed rate (FRM) ,,4,4,,,2,2,1, households to service their loans except for those who have seen their payments rise significantly Thirdly, while the delinquency rates for both subprime and prime ARMs have risen, it is not much higher than the average seen during the late s The recent rise in delinquency rates is entirely driven by ARM loans, which now have a larger aggregate impact than earlier due to their higher market share Rise in total delinquencies driven by ARMs %-point %-point 6 Total delinquency rate (MBA sourced) 4 other 2 Contribution from FRM ARM Macroeconomic ramifications While the subprime mortgage sector clearly faces some serious problems, the magnitude of the macroeconomic spillover might not be that big After all, the subprime market constitutes a relatively small share of the total mortgage market and indeed the economy However, we look at some potential macroeconomic effects below: - Subprime problems could indicate or facilitate a broader tightening of credit standards for consumers as well as corporates Several interesting firsthand observations are deductible from the delinquency and foreclosure data Firstly, the rise in delinquencies and foreclosures is visible in both the prime and the subprime segments, but only in ARM loans This suggests that the problems are basically related to the type of loan (ARMs) and not the type of borrower Moreover, it is unsurprising that ARM-borrower performance should deteriorate in the wake of 42bp of policy tightening Secondly, there has been no rise in delinquencies or foreclosures in FRMs in either the prime or the subprime segments This is especially interesting with regard to the subprime sector, as delinquency and foreclosure rates for ARM and FRM loans are usually closely linked Moreover, the fact that FRMs have seen no rise in delinquencies suggests that there is no broad deterioration in the ability of US - The pick-up in foreclosure and delinquency rates could delay the stabilisation of the housing market by putting more homes up for sale on the market - Foreclosure and delinquency data indicate that problems are primarily concentrated in ARM loans Hence, it is of interest to look at the impact of ARM resets on both prime and subprime borrowers Risk of a broad tightening in credit standards One possible risk from the problems in the subprime market is a spillover in terms of a broader tightening of credit standards both in the prime mortgage market and with respect to other types of consumer or corporate loans Consumer credit growth outside housing shows, so far, no sign of a broader tightening in credit avail- DANSKE BANK

4 RESEARCH MARCH, 2 ability Recessions have usually been accompanied by a major drop in credit growth from very high levels There are at present no such distress signs Consumer credit growth has generally been stable and very modest during the expansion and remains so No sign of stress in consumer credit growth 2 % y/y % y/y 1 Consumer credit The recent Senior Loan Officer Opinion Survey conducted by the US Federal Reserve supports this picture and shows no signs of a broader tightening in consumer credit standards outside the mortgage market While a tightening cannot, of course, be ruled out in the coming quarters, the historical pattern does not suggest any strong correlation between a revision of credit standards in the mortgage sector and a revision in the other consumer loans included in the survey Tighter credit standards for mortgages, but not for other consumer loan types % of respondents % respondents 6 Senior loan officer opinion survey tighter credit standards 4 Credit card Other consumer loans 2 - Mortgage loans Moreover, the availability of credit in the corporate sector does not show any alarming patterns Credit standards have simply been tightened gently during recent quarters as a response to Fed policy tightening No alarm bells in corporate credit availability - % of respondents % of respondents Senior loan officer opinion survey tighter credit standards small C&I loans Fed funds - large and medium Corporate credit spreads are close to the level in 2 and 26 (see chart below) This is supported by the very strong fundamentals in the corporate sector, with corporate savings remaining unusually high in this expansion Credit spreads remain low bp bp BBB A AA AAA Generally speaking there are no signs of a broader spillover to either the prime mortgage market or non-mortgage consumer and corporate credit That said, the potential, timing and evolution of any broader credit tightening is difficult to gauge However, with consumer and corporate fundamentals remaining healthy (ie low unemployment, high income growth and strong corporate profits and balance sheets), the conditions for the economy to enter a vicious circle that would lead to a broader credit meltdown are really not present Credit standards will, though, probably tighten somewhat during the coming quarters partly as a lagged response to the tightening of Fed monetary policy and partly as a reflection of increased prudence from lenders in the wake of the subprime problems Prolongment of the housing downturn Another potential fallout from the subprime woes is that higher rates of delinquency will mean more foreclosures and hence upward pressure on the inventory of homes for sale According to the American Housing Survey conducted by Census in 2, around one-third of all US households have no mortgage debt Hence, a rise in delinquencies will affect only two-thirds of the roughly million owner-occupied homes in the US For each 1 age point increase in the foreclosure rate, approximately, additional homes will be put up for sale each quarter, or 2, on an annual basis DANSKE BANK 4

5 RESEARCH MARCH, 2 Higher foreclosure rates mean more homes for sale % % Foreclosure rate all homes (MBA sourced) Since Q1 26 the overall foreclosure rate has increased from 41% to 4% (up 12 age points), which is an all-time high (note, though, that this seems worse than it is in reality, as there is a persistent upward trend in the foreclosure rate) The 12 age point rise in the foreclosure rate has, or will, put 2, more homes on the market However, as the current rise in delinquencies promises a further increase in foreclosures, assuming another 12 age point increase in 2 seems reasonable Hence there would be, additional homes for sale compared to Q1 26 This corresponds to less than 6% of current annual total home sales or a little less than one months turnover This would add approximately 8 age points to the monthly supply of total homes (the inventory:sales ratio) In a simple regression model on real OFHEO house prices (modelled on lagged real house price growth, change in interest rates, unemployment and the inventory:sales ratio), the partial effect would be a 4 age point lower appreciation in house prices In other words, while the rise in foreclosure rates will have some negative impact, it will be minor both in terms of house price growth and relative to the total turnover in the housing market For sure this is a negative factor for the stabilisation of the housing market, but again the direct impact is minor Moreover, some of the rise in foreclosures in 26 has probably already fed into the housing market, contributing to some of the current downward pressure on house prices Impact of ARM resets The delinquency and foreclosure data reviewed above indicate that the problems in the mortgage market are closely tied to the resetting of ARM mortgage rates This suggests there might be some macroeconomic impact, as rising ARM rates will eat up income otherwise intended for consumer spending ARM interest rates going up % % Average ARM constract rate % 4 ARM 1-year contract rate % Given that approximately 2% of mortgage loans are ARMs, the home ownership rate is 6% and only two-thirds of US homeowners have mortgage debt, this will impact only % of US households a relatively limited age As the exact profile of ARM resets is difficult to obtain, it may be useful to begin with a worst-case calculation to asses the maximum potential of the problem To do this, we make the following assumptions: 1) All ARM were fixed when the ARM rate was lowest, in autumn 2, and the entire amount will be reset in 2 2) All ARMs are 1-year ARMs These are the ones that have seen the greatest increase in the contract interest rate since late 2, ie close to bp ) No cheaper financing alternative exists 4) Disregard the fact that there is a 2bp cap on many ARM loans, implying that the ARM contract rate cannot jump more than 2 age points at each reset ) Use the OFHEO estimate of total outstanding one-family ARMs of USD 24bn in Q2 26 This scenario is illustrated in the table below The effect on consumption corresponds to a age point reduction in nominal income growth Worst-case scenario for ARM resets Outstanding single-family ARM (OFHEO Q2 estimate) 24 billion USD Increase in contract interest rate bp Increase in ann mortgage payment 4 billion USD Nominal pers income in billion USD Forecasted growth in pers nominal inc 2 6 % Nominal pers income in 2 1 billion USD Nominal pers income adj for reset cost 6 billion USD Nominal pers income growth adj for reset costs % Reset cost in terms of nominal income growth %-point DANSKE BANK

6 RESEARCH MARCH, 2 While the impact of this (worst-case) scenario is definitely a significant shock to consumers (especially as it is likely to hit the most credit constrained consumers), it not an insurmountable shock from an aggregated perspective It corresponds to a reduction in nominal income growth of USD 4bn Put differently, the GDP effect of the worst-case ARM reset scenario is comparable to a USD oil price shock, according to our consumption multiplier Below, we have estimated another more realistic scenario where ARMs will be gradually reset during the period 26 to 2, with the bulk of resets coming in 2 and 28 We have kept the rest of our conservative assumptions While the profile is not exact, it is well within the range of circulating estimates ARMs to be reset Impact of ARM resets (26-2) Value of ARMs to be reset Increase in annual mort Payment Nominal pers Income Reset cost % USD bl USD bl USD bl % y/y % pers inc % % % 2% % 24% % 6% % % Again, this does not give reason for any major concern ARM resets will subtract 2 age points from nominal income when the effect peaks in 2 and 28 And again note that these estimates are derived from rather harsh assumptions about interest rate fixings, caps and alternative financing options Hence, an effect corresponding to less than 2% of nominal incomes would probably be close to what plays out Given robust income growth of close to 6% and very sound consumer balance sheets, the impact of the ARM resets does not appear to present a serious shock to the consumer from an aggregate perspective ARM resets in a broader perspective One problem with the analysis of the ARM resets above is that it does not tell the whole story For instance, it does not take into account the change in cashflow from other items on the consumer balance sheet In the table below we have depicted assets and liabilities from the Flow of Funds consumer balance sheet, where the cashflow is either fully or partly dependent on movements in short-term interest rates Asset and liabilities with (short-term) interest rate related cashflow Assets ARM mortgages billion USD 42 Deposits 66 Open market paper 188 Security credit 66 Sum 6 Liabilities ARM mortgages 242 Bank loans 6 Consumer credits 248 Security credit 22 Sum 2 Net 21 Note: ARM mortgages are calculated from an assumption that 2% of the value of all mortgages is placed in ARMs This is consistent with OFHOE numbers from 26 Q2 Interestingly, the stock of assets with an interestrate-sensitive cashflow is larger than the liabilities This actually implies that the net interest income (or interest-rate-related cashflow) will rise along with the short-term interest rate Hence, from an aggregate perspective, the negative cashflow effect from the ARM resets is more than countered by the larger cashflow on interest-ratepaying assets This effect is illustrated in the chart below Cashflow effect from higher interest rates % - point % 2 << Chg in net interest income 2 in of disp income year interest rate >> This is not to say that higher interest rates are not having a contractive effect Interest rates of course work through several other channels (ie asset valuation, credits, etc) and for sure higher rates have a net negative effect on the economy Moreover, there might be an asymmetric effect from the changes in cashflows However, the point is that one cannot judge the negative impact from ARM resets without taking into account the effects of higher interest rates on the entire consumer balance sheet 2 DANSKE BANK 6

7 RESEARCH MARCH, 2 Conclusion There is no doubt that the problems in the subprime mortgage market are serious Potentially they could be an early warning of more widespread troubles in the credit market, with a negative macroeconomic spillover While signs of such developments should be watched closely, we expect the problems to remain contained within the subprime sector, which has the largest share of ARMs If this is the case, the ramifications for the rest of the economy will only be minor Moreover, the direct impact of ARM resets and rising foreclosures will be negative but not disruptive In general it might be constructive to view the subprime woes as a symptom rather than a cause After some 42bp of monetary tightening and given that the housing market is slowing, it is understandable that some problems show up in the riskiest area of the mortgage market While we would not ignore the risk from the subprime mortgage sector, we are not overly concerned about the direct impacts The bottom line is that with the economy remaining sound outside the housing market, the conditions for a broader credit meltdown do not seem to be present Hence, given the current situation we do not expect any serious ramifications for the macro economy That said, it is important to closely watch for any signs of a broader and more severe tightening of credit standards in the coming quarters This report has been prepared by Danske Research, which is part of Danske Markets, a division of Danske Bank Danske Bank is under supervision by the Danish Financial Supervisory Authority Danske Bank has established procedures to prevent conflicts of interest and to ensure the provision of high quality research based on research objectivity and independence These procedures are documented in the Danske Bank Research Policy Employees within the Danske Bank Research Departments have been instructed that any request that might impair the objectivity and independence of research shall be referred to Research Management and to the Compliance Officer Danske Bank Research departments are organised independently from and do not report to other Danske Bank business areas Research analysts are remunerated in part based on the over-all profitability of Danske Bank, which includes investment banking revenues, but do not receive bonuses or other remuneration linked to specific corporate finance or dept capital transactions Danske Bank research reports are prepared in accordance with the Danish Society of Investment Professionals Ethical rules and the Recommendations of the Danish Securities Dealers Association Calculations and presentations in this report are based on standard econometric tools and methodology Documentation can be obtained from the above named authors upon request Major risks connected with recommendations or opinions in this report, including as sensitivity analysis of relevant assumptions, are stated throughout the text First date of publication Please see the front page of this research report This publication has been prepared by Danske Bank for information purposes only It is not an offer or solicitation of any offer to purchase or sell any financial instrument Whilst reasonable care has been taken to ensure that its contents are not untrue or misleading, no representation is made as to its accuracy or completeness and no liability is accepted for any loss arising from reliance on it Danske Bank, its affiliates or staff, may perform services for, solicit business from, hold long or short positions in, or otherwise be interested in the investments (including derivatives), of any issuer mentioned herein Danske Banks research analysts are not permitted to invest in securities under coverage in their research sector This publication is not intended for private customers in the UK or any person in the US Danske Bank A/S is regulated by the FSA for the conduct of designated investment business in the UK and is a member of the London Stock Exchange Copyright ( ) Danske Bank A/S All rights reserved This publication is protected by copyright and may not be reproduced in whole or in part without permission DANSKE BANK