This is essentially a rerun of this earlier post, which produced no answers I found convincing. It's provoked by Livio's recent post, which produced comments raising questions similar to one that I've been asking for many months now. The question is: what are the mechanics behind "there's a housing bubble in Vancouver" (it's always Vancouver) to "therefore, Canada is doomed to a US-style balance-sheet recession"? There are any number of pundits who are willing to make this jump in logic, but the intermediate steps are almost never enumerated.
I don't have any problem imagining that housing prices in Vancouver and elsewhere will fall when the Bank of Canada starts its next cycle of interest rate hikes. What I'm having problems seeing is why that means we will follow the path the US took.
Of course, before we get too concerned with the power of the Vancouver housing market over the Canadian economy, we have to answer the question: "What is so special about Vancouver?" After a sharp increase during the mid-2000s, Calgary prices peaked in mid-2007, and fell more than 15% before bottoming out in 2009. They have yet to recover their peak. Why is a fall in Vancouver housing prices a national crisis to be dreaded, while a fall in Calgary housing prices is not?
Anyway. Brad DeLong claims that
In order to have successfully predicted that we would be where [the US is] now, you would have to have predicted a large number of things:
-
That a global savings glut and a period of low interest rates would produce a housing boom.
-
That the housing boom would turn into a housing bubble.
-
That the housing bubble would lead to a collapse of mortgage underwriting standards.
-
That risk management practices on Wall Street would have been nonexistent.
-
That the Federal Reserve would not be able to construct its usual firewall between finance and the real economy.
He continues on, but says that the most imaginative got to point 5 at most. Canada is running a current account deficit these days, so point 1 can't be dismissed out of hand. Point 2 is a definitely a possibility – but if all the bubble stories are datelined Vancouver, then it's hard to see how much of an effect it will have on the rest of Canada.
Point 3 seems to be where we have to stop the analogy to Canada. After a brief flirtation with longer amortisation periods and zero down payments – which is as far as we got to reproducing the subprime excesses of the US – Canadian regulators have spent the last few years tightening mortgage requirements. Is there any sign that Canadian mortgage underwriters have adopted pre-crisis US-style practices? The Canadian business press has been very keen to match the horror stories coming out of the US, but no amount of digging seems to have produced anecdotes corresponding to the 'liar loans', 'teaser rates' and 'NINJA loans' that were the standard fare of the pre-crisis US financial press.
The CMHC's short-lived and long-dead experiments pale in comparison with such exotica. It's not at all clear to me that an eventual weakening in prices will bring about a massive entry of houses on the supply side of the market. Who would be selling those houses, and why? And if there is no sudden influx of houses on the market, why would prices crash? It seems more likely to me that house prices would stay stable or decline slowly as nominal incomes grew to levels compatible with existing home prices.
Point 4 also seems moot – Canadian mortgages haven't been securitised in the bewilderingly opaque way that pre-crisis US mortgages were. If there were to be a massive wave of Canadian mortgage defaults, there wouldn't be much mystery about who would be on the hook: it would be the CMHC, and, by extension, the federal government. I don't see why financial markets would seize up in a fit of counterparty confusion the way Wall Street did in the fall of 2008.
Moreover, the odds of a massive wave of Canadian mortgage defaults seem small. As everyone know by now, Canadian mortgages are generally recourse, so the incentives to walk away from homes that are underwater are greatly reduced. If Canadians refused to default in large numbers when mortgage rates went above 20% in the early 1980s, why would they do so if they go up a couple of points from where they are now?
I'd like to think that point 5 is also moot, to the extent that the Bank of Canada is – so far – more insulated from partisan politics than the Fed is. And of course, the Bank can also learn from the Fed's mistakes.
But I may be missing something in the roadmap that takes us from here to housing market Armageddon. If so, could someone tell me what it is?
Worthwhile Canadian Initiative 
I’ll outsource to some people who have done a lot of empiricial research into the CRA’s role in the housing crisis:
Barry Ritholtz (Big Picture) http://www.ritholtz.com/blog/2009/06/cra-thought-experiment
Federal Reserve Governor Randall Krozner: http://www.federalreserve.gov/newsevents/speech/kroszner20081203a.htm
SF Fed Economists: http://www.richmondfed.org/conferences_and_events/research/2008/pdf/lending_in_low_and_moderate_income_neighborhoods.pdf
If you read the above (particularly the Fed pieces) and are still convinced the CRA is to blame, then there is no hope for you.
This is from Ian Lee. (He tried to post, but it refused to accept it. That happens to me sometimes. I log out, then log back in again.)
Agreed, there have been several Fed publications arguing as you, Ritholz and Krugman argued, that the CRA did not cause or motivate or ensure that any sub prime mortgages were approved due to the CRA.
So lets return to first principles & the CRA itself & then move forward with some questions – not answers.
Philadelphia Fed, 2008: “Enacted by Congress in 1977, the CRA requires bank regulators to encourage insured depository institutions, to help meet the credit needs of their entire community, including low- and moderate-income areas. The CRA requires federally regulated and insured financial institutions to show they are lending and investing throughout their assessment areas, which are defined by the banks as areas in which they accept deposits and make a majority of their loans.[3] One of the main principles behind the CRA is that banks and thrifts benefit from the deposits of low- and moderate-income households; in return, they should open access to credit in these communities.
From the Dallas Fed, 2009: “By opening access, the CRA enables creditworthy low- and moderate-income individuals to become part of the financial mainstream. Since its passage, the CRA has leveraged an estimated $4.5 trillion in these communities and helped to create jobs, develop small businesses and make mortgages accessible.[4]
Ian’s synthesis: The CRA was first passed (without regulatory teeth) in the Carter Administration in 1977, in an attempt to stop banks from “redlining”. It was amended – with regulatory teeth – in 1996 during the Clinton Administration.
What banks? Commercial banks (not investment banks) as the 6,000 commercial banks across the US grant mortgage loans, consumer loans and small business loans.
Banks try – using databases, credit scoring, job checks, credit checks etc. – to lend money to people who will repay the loan. Inter alia, practices of redlining developed of refusing to lend to people from poor communities e.g. high percentage of renters, low average income, seasonal and high employment et al.
(I redlined all the time when I was in the bank. We would not lend to people on welfare or those living in subsidized housing projects or in Vanier which in the 1970s was known for drugs and other high risk activities).
The not CRA critics fully acknowledge that yes, the CRA was designed to stop these bad, horrible practices (of using risk metrics and proxies to minimize losses).
BUT critics implicitly argue the CRA was an abject failure – because it did not cause or influence any bank anywhere in the US to make any mortgage loans to any low income borrower in low income neighborhoods – the express intent of the CRA.
Stop for a moment. For the better part of 100 years, American commercial banks approved billions of dollars of mortgages using due diligence & credit underwriting standards developed during that hundred years.
Yet, sometime in the late 1990s – after the passage of the CRA with teeth, several thousand banks across the US suddenly threw out all due diligence and standard underwriting policies and procedures of a hundred years.
And for the first time, commercial banks started to securitize a majority of the mortgage volumes booked – up dramatically from less than 1/3 before 1996 to over 2/3s in the early nought decade. BUT it was not – repeat not – because of the CRA. Not at all. So what caused this change?
1. George W Bush and 2. greedy corrupt banks
In the late 1960s, I used to watch a TV show – Rowan and Martin’s Laugh in – with Goldie Hawn and Flip Wilson. Wilson had a character called Geraldine Jones who whenever she was caught doing something bad, said in a high pitched falsetto voice, “the devil made me do it”.
In our bucolic innocent past pre 2001 George W Bush, banks were not self interested i.e. “greedy”. Greed was a new phenomenon brought by the “devil” himself – George W Bush. This is the “religious” nonsense of the left that is just as fatuous as creationism and similar nostrums on the right.
When a critic provides a credible empirically grounded hypothesis to refute the very substantial empirical research of Calomiris and Leibowitz and the “right wing meme” i.e. the New York Times and Getchen Morgenson – I will pay very close attention. To this point, I have not reviewed a more robust hypothesis – other than the negative “it was not the CRA”.
The above is from Ian Lee.
Last post, because we are already at the point of talking past each other – but even Gretchen Morgenson and Josh Rosner don’t blame the CRA: http://economistsview.typepad.com/economistsview/2011/07/we-do-not-blame-cra.html
Man, where to start with this one?
First the guy cops to prejudice in lending.
Next he invents a completely fictitious argument to attack. Can he name anybody who made that argument? Conceivable, but nobody I know.
The history of the subprime disaster is very clear. One key was the originate and sell model for mortgage lenders. Once mortgage lenders could sell all of their mortgages they no longer had any constraints on making loans to unqualified buyers. The other key was the ratings failures on Collaterized Debt Obligations. By combining bad debt in opaque ways the merchant banks scammed the agencies, which led to them being able to scam their customers. Ultimately they scammed themselves.
The reason it happened when it did had more to do with financial inventions than the CRA.
At peak 25% of US mortgages were subprime. Did the US banking system really redline 25% of the country?
Last post, because we are already at the point of talking past each other – but even Gretchen Morgenson and Josh Rosner don’t blame the CRA: http://economistsview.typepad.com/economistsview/2011/07/we-do-not-blame-cra.html
Possibly so. Thanks for the reference, as I had not read this piece.
So – now to clarify what I am arguing.
1. Agreed – the Dems did not cause the subprime crisis – for EVERY President since Roosevelt – Dem and Republican – in the annual state of the union address, said to the effect, “it is the right of every American to own a home”.
2. the CRA was merely a symbol and a symptom of interference by the US Congress – by both parties – in the housing market. While we should all want vigorous regulatory supervision of all financial institutions – including shadow banking and not just banks – it is massive overreach for govt to contemplate second guessing lending or investment decisions down to a zip code or postal code.
And this brings me to Jim’s response who was shocked (like Captain Renault in Casablanca to discover gambling in Rick’s casino) to discover that banks “discriminate – with prejudice” against high risk customers.
Yes – and auto insurance firms charge high insurance premiums to young men and life insurance firms will not insure a pre existing condition and if you declare bankruptcy, it is very difficult to obtain credit … Indeed! As they say in Quebec, “mais voyons en donc!”
Bank managers are promoted – for a number of reasons – including importantly, achieving or exceeding loan growth targets but also achieving delinquency targets. Decisions to reject applicants due to low credit score or low income including welfare income or inadequate assets (assured living in subsidized housing), are based on explicit policies and implicit understandings(i.e. what Polanyi called tacit knowledge) concerning risk characteristics.
The CRA was passed and then amended and as several Fed “branch” research depts noted, the CRA ie Congress, attempted to pressure or require lenders to provide lending monies to lower income people in lower income communities.
And, we do know that large numbers of subprime mortgages were approved from late 1990s to 2008(data in my presentation) and later defaulted.
The fed and others reject that the CRA achieved what it was supposed to achieve. Fair enough.
However, a sub prime crisis did unfold with large numbers of banks throwing their due diligence and underwriting standards out the window.
What hypothesis can logically and empirically explain this?
Jim suggested “innovation” i.e. of securitization. Given that securitization was “innovated” decades before the meltdown in 2008. Indeed, Fanny was created in 1938 to promote and develop a secondary market in mortgages – a precursor to securitization. (at BMO in late 1970s, we sold privately placed securitization bundles of about 10% of our mortgages written annually to pension and life insurance firms).
And we know that American commercial banks granted mortgages for about a century without the subprime problem – in other words they possessed the competencies to approve sound mortgages that did not default.
My hypothesis was a change in public policy – manifested inter alia but not limited exclusively to – the CRA.
This “govt policy intervention” hypothesis – that the US Congress intervened on bi-partisan level to push or require banks to lend to low income people in poor areas – advocated by Prof Leibowitz and Prof Calomiris amongst others, was rejected.
Suggest a more compelling hypothesis supported empirically.
Jim – one more quick point on the innovation hypothesis, as I have come across this explanation on several occasions in the last 3 years, to explain the collapse in underwriting standards and thus the sub prime crisis.
When we securitized a percentage of our mortgages annually (we knew far in advance that it would occur as it was a policy) we were often nervous, as we thought, “what if one of the mortgages is a lemon that goes delinquent – they i.e. Sun life, will think we are trying to stiff them:”. IF you want repeat business, you are not going to stiff your customer.
Securitization should – ceteris paribus – ensure HIGHER underwriting standards – not lower credit standards because when the customer eg investment bank or pension fund discovers that you sold them garbage, there will be serious push back and a loss of a customer. That is a negative outcome for a career banker.
What is the logical relationship between financial innovation producing securitization and lower underwriting standards?
What explains the abandonment of years of practice, experience and knowledge by several hundred thousand career mortgage managers across the US?
And given that the Fed acknowledged the express intent of the CRA in several research publications, if indeed it was a complete failure across the US in achieving what the CRA was supposed to do, why was the act a failure? i.e. why did the bank inspectors not enforce what they were required to enforce in the 1996 CRA? Who was guarding the guards?
These are the questions that are not being asked or addressed.
Ian:
Did you give equal consideration to prepayment risk? My virtue of exercising the prepayment option on a mortgage deprives the bank of interest. In basic finance books this is explicitly stated as a characteristic risk of mortgage securities vs. government bonds.
Ian: I don’t know anything about CRA specifically. Based solely on what you’ve claimed above – that ‘CRA with teeth’ forced banks to lend to people they wouldn’t otherwise have lent to – then I would expect that there are documented cases of regulators taking action or sanctioning banks in some way. Did this in fact happen? Are their documented cases of litigation (for example)?
Off the cuff, I’d say the causation ran the other way: bankers wanted something, anything to wrap in glossy prospectus and stamp AAA on, and any regulator who dared get in the way was run out of town (see Born, Brooksley).
Determinant – prepayment risk was and is always there, which is why banks in my time in banking (70s and early 80s) charged a 6 month prepayment penalty (which was only waived if the mortgage was being renewed for a larger amount or due to foreclosure because under the relevant legialtion, once in foreclosure, a lender cannot recover prepayment penalty).
Based on anecdotal evidence today, it is my understanding that the Cdn banks today are charging larger amounts of prepayment penalties to compensate for the lost income when the mortgage is prepaid (the prepayment risk you noted). But that risk has always been present – typically excercized by professionals who move from one city to another necessitating the sale of the home.
Patrick – you ask good questions.
Leibowitz in his most recent research found: “The analysis [of McDash analytics with over 30 million mortgages] indicates that, by far, the most important factor related to foreclosures is the extent to which the homeowner now has or ever had positive equity in a home. The accompanying figure shows how important negative equity or a low Loan-To-Value ratio is in explaining foreclosures (homes in foreclosure during December of 2008 generally entered foreclosure in the second half of 2008). A simple statistic can help make the point: although only 12% of homes had negative equity, they comprised 47% of all foreclosures”.
“To be sure, many other variables — such as FICO scores (a measure of creditworthiness), income levels, unemployment rates and whether the house was purchased for speculation — are related to foreclosures. But liar loans and loans with initial teaser rates had virtually no impact on foreclosures, in spite of the dubious nature of these financial instruments”.
Source: “New Evidence on the Foreclosure Crisis: Zero money down, not subprime loans, led to the mortgage meltdown”, July 3, 2009, WSJ
Re your comment: “I’d say the causation ran the other way: bankers wanted something, anything to wrap in glossy prospectus and stamp AAA on it”.
Yet, although securitization had been around for decades, the motive you hypothesize did not occur previous to the late 1990s. Moreover, as I noted above, why would any business manager in any credible (not fly by night) business want to shaft his customers, thereby ensuring zero repeat business thereby contradicting the self-interest argument.
The causal question remains: if the collapse of mortgage underwriting standards was not caused by a change in US Govt policy manifested in the CRA (and related policies of HUD, Fanny et al), what caused the change in behaviour of century old mortgage underwriting practices? What was the causal factor that caused the 180 degree in underwriting in the late 1990s?
A really clever research project would be to examine the mortgages written from say 1998-2008 by US commercial banks and divide them into two populations: the 2/3 that were securitzed and the 1/3 that were retained on the books. Was there a differnece in eg down payment, credit history, mortgage default rates?
Another research set – to try to answer Patrick – would be to examine the bank inspector reports from before 1996 to after 1996 to see if references in these bank inspection reports were made to the duty imposed on banks (per the CRA) to lend to low income people in low income neighborhoods. If so, how was it measured? using census income data by zip code? and waht were the thrreats – if any – recommended or imposed by the bank inspectors.
Last point – the Dallas Fed in 2009 said:
“By opening access, the CRA enables creditworthy low- and moderate-income individuals to become part of the financial mainstream. Since its passage, the CRA has leveraged an estimated $4.5 trillion in these communities and helped to create jobs, develop small businesses and make mortgages accessible.[4]
Did not the Fed, Krugman, Ritholtz et al agree that mortgages were NOT granted because of the CRA – what then is the “$4.5 trillion in these communities … to make mortgages accessible”?
Ian: CRA is not something I have researched in any depth. Not my country, and not my area. My sense is that CRA was bad policy, but I think that things would have unfolded in much the same way even without the CRA. Other countries (England, Ireland, etc.) had similar problems but no CRA. Thoughts?
Nick – yes, I agree completely.
It brings me back to an earlier comments I made. The CRA is merely a symbol or an expression of the intent of policy makers to attempt to micromanage lending down to a zip code, without understanding the multiple variables that factor into a business loan or mortgage credit decision. I am not trying to suggest that the credit decision is mysterious and unfathomable. Rather, it is analogous to the grading system in university Appeal Committees (I sat for the last 3 years on the Grad Appeal Cmte – which acts prudently concerning second guessing the professor being appealed, for that professor possesses vastly more contextual knowledge concerning the student from his/her class behaviour).
If the US Congress had been worried about racism as the underlying social ill, they could rely on the Civil Rights Act and the plethora of anti racism statutes passed, to punish behaviour of this type.
But to return to your larger point, yes, any country can have a similar mortgage finance crisis without a CRA. All a country need do is reduce the minimum required down payment to obtain a mortgage – for as Stan Liebowitz showed in his empirical research and I know from banking data and my own previous banking experience, the single most important predictor of mortgage default is low or no equity.
Last year, I testified before the House of Commons Finance Cmte and urged them to recommend increasing the minimum down payment for a high ratio mortgage to 10% and to recommend reducing the amortization from 35 years to 30 years maximum. I was also invited to one of the Finance Minister’s pre budget consultations and I recommended the same policies.
Needless to say, the housing development industry and the real estate brokers vehemently disagreed, for it would significantly reduce sales they argued. I cheerfully agreed that it would indeed decrease sales, as it would increase the “barriers to entry”, i.e. the threshold to owning a home.
Ironically, anti-poverty groups agreed with the developers and brokers as increasing the DP would involve, they argued, “discrimination against low income people”. To which I responded, the only person who has the “right to own a home” is the person with the required down payment who can afford the monthly PIT payment AND monthly utilities AND ongoing maintenance of said home. Yes, indeed, it is true that not everyone can afford to own a home. But then not everyone can afford to own a Porsche or a Volvo – like me (as I am a poor professor!).
But this brings us to the front door of the CRA or equivalent statutes of social engineering that try to gerrymander the financial rules.
I wished instead, these advocates would lobby for a National Securities Regulator or to bring provincially regulated pensions under the exclusive regulatory supervision of the Govt of Canada so that capital markets, deposit taking institutions, pensions and all shadow banking would be regulated by one federal authority.