« Sovereign CDS -- Random Thoughts | Main | Bankruptcies Maintain Similar Month-to-Month Rate in January »

Monetary Policy and the Housing Bubble

posted by Adam Levitin

A popular explanation of the financial crisis lays the blame at the feet of the Federal Reserve for lax monetary policy.  In this story, the Fed dropped interest rates starting in 2001 and kept rates too low for too long.  Low rates induced an orgy of mortgage borrowing for leveraged home speculation. 

It's a nice story.  Only problem is it doesn't really hold up under inspection.  Low rates in 2001-2003 did fuel an amazing mortgage refinancing boom, but not a purchase boom, and the boom was mainly in conventional fixed-rate mortgages, not the exotic products later years.  Moreover, despite the refinancing boom, no housing bubble was emerging in this period. 

The Fed started to raise rates in mid 2004 and continued to do so until mid-2006.  It was during this period that the bubble emerged, when rates were going up.  (To be fair, some might argue for an earlier date to the bubble, even as far back as the late 1990s.)  If we date the bubble from 2004, it's not consistent with a rate-driven bubble story, although rates were still extremely low in absolute terms during this period. 

The monetary policy story, however, really falls apart when one compares the US and Canada, as the graph below does.  Canadian interest rates, and perhaps more importantly, Canadian mortgage rates, track US rates pretty closely.  Yet the US had a housing bubble, and Canada did not.   This means we have to look somewhere other than monetary policy to explain the housing bubble.  The answer, I believe, lies in method and regulation of housing finance. 

US Canadian Mortgage Comparison

Canada has a very different system than the US.  There is almost no private-label securitization in Canada; mortgages are either on banks' books or securitized in issues that "bear the Beaver."  (Ginnie Maes, guaranteed by the US government, are said to "carry the Eagle," and different types of FHA-approved lenders are referred to as Full Eagle or Mini Eagle lenders.) 

Canadian Residential Mortgage Holdings

Moreover, all mortgages with LTV>80% are required to have private mortgage insurance, and there's a pretty limited market, basically the Canadian Mortgage Housing Corporation (government entity) and Genworth.  The rates are identical (perhaps regulated), so the only way to gain market share is to loosen underwriting, and that's a riskier move for an insurer to make (yes, they can reinsure, etc. but still). 

The Canadian system seems to have been able to keep out most of the exotic products that sprouted up in the US.  There is concern that Canada is currently developing a housing bubble (their price/rent ratio is pretty high).  Indeed, the growth of NHA MBS makes one wonder if all the questionable mortgages are just getting picked up by the government, akin to what might be happening here with FHA now.  Nonetheless, it seems that so far, the Canadian system likely shielded our northern neighbors a mortgage market disaster. 

This isn't to say that the Fed was blameless in the housing bubble.  But the Fed's mistakes don't seem to be in monetary policy as much as in regulation, such as the failure to act on its HOEPA regulatory authority until after the crisis occurred.


Comments

I'm hoping we'll get some local color in the comments. Pottow, perhaps?

Got some of them fancy colored graph things for the U.S. Residential Mortgage Holdings for the same time period?

Huh? The housing price indices were all going up in the early 00's. rising prices lured suckers into the market. And of course there was also a failure of regulation.

The analysis is a bit superficial.

There were many second order effects to the "Greenspan Put", not the least of which was the consensus view that nominal house prices could never decline as long as the Fed was ready to aggressively cut rates.

Leaving the "Put" aside, however, there was a more tangible second-order effect. The refi boom resulted in a huge increase in origination -- I believe mortgage volume topped $2.8tr in 2003, compared to half that a few years earlier. The practical consequence was that mortgage originators faced earnings declines in the ensuing years. In an effort to show earnings growth, the likes of New Century, Indymac, Ameriquest, Merrill, Bear, WAMU and Countrywide decided on a two-pronged strategy: 1) ease standards to raise volumes AND margins (no-doc lending exploded); and 2) aggressively expand the balance sheet (i.e. increase risk) to earn more "spread". Both of these decisions led directly to the financial crisis. You may call this result a "failure of regulation". Fine. That failure of regulation, however, would never have happened but for the "Greenspan Put".

Finally, how do you think "regulatory capture" happened? Greenspan was the de facto "Chief Economist" of the U.S. Which regulatory agency head had enough clout to go against him? You see, Indymac argued with regulators that their "models" showed that nominal house prices never declined, and Greenspan basically argued the same. So, if you went against no-doc lending, it was for one reason and one reason only: because you thought house prices COULD decline. As a regulator, that thought would put you at odds with Greenspan and the universal belief in the "Put". That was simply not going to happen in 2003-2006.

First, no one denies that there are other factors involved. Regulations clearly are up there with monetary policy as one of the most significant factors. Booms normally manifest in specific sectors due to completely different reasons. There is no reason why only a housing boom is the inevitable result of too easy monetary policy. It could be a surge in some other form of investment (maybe the boom in Canada was in commodities investment?). I don't know of any serious thinker who supports the easy monetary policy as a source of the housing boom who would disagree with you that regulations are the reason why there wasn't as significant a boom in Canada. However, they still say that it was a factor in the U.S.

Second, when people make the argument that rates were too low, they can talk about in an absolute sense or a relative sense. While theory might dictate that rates should be thought of in terms of relative to some natural rate of interest, it is impossible to observe the natural rate of interest. Hence, I look at the short-term interest rates relative to what the Taylor rule proscribes as a first approximation of how tight/loose monetary policy is. On this basis, the U.S. policy was very easy in 2004-2006. I don't have access to Factset on my home computer and don't feel like searching online to get the Canada data (perhaps someone else can?), but I would guess that relative to a Canadian Taylor rule, the BoC policy wasn't as easy as the Fed's was.

Third, the reason why easy monetary policy impacted the U.S. housing market is because it brought down short-term rates which were used in ARMs. The percentage of mortgages that were ARMs rose sharply over this period as they became more attractive with the yield curve steeply sloping upward. This fueled demand for housing, putting upward pressure on prices, and then we were off the races. In Canada, if the typical home is financed by 5 year FRMs amortized over 25 years (according to the IMF), then there isn't as much room for low short-term rates to impact this market. However, the share of ARMs seems to be growing, so Canada might be more sensitive in the future. The key determination for Canada is then the long-term rates that are determined by the market. So I'm not surprised at all that Canada wouldn't have a housing bubble.

Finally, the easy money story gets us to the housing bubble, but it doesn't get us to a financial collapse. For this, you have to look to other factors and I don't deny that. Banks were overleveraged and it's not like you can ignore that. When it comes to the explosion of exotic mortgages or securitization, you have to fault the rating agencies failing to adequately measure the risk of the securitized instruments, regulations like Basel 2, and poor lending standards (which I attribute to eagerness to take advantage of the housing boom). Really only one of those factors is related to easy money. A thorough investigation of this crisis cannot ignore these other factors. So while easy monetary policy is one important driver of this crisis, it is by no means the only thing that you have to look at.

Price indices aren't rigidly representative. But in order to refinance or purchase property there must be a (1) turnover of a (2) specific volume of money.

Both the transactions velocity, and our means-of-payment money, are regulated thru the FED's mandate (the one that requires price stability).

Monetary policy objectives are formulated in terms of desired rates-of-change (roc's) in monetary flows (MVt) relative to roc’s in real GDP.

While specific prices aren't targeted, housing & commercial construction as a percentage of real-output had to be (I'm guessing), historically skewed relative to real-output. So obviously the FOMC needed a much more tighter monetary policy.

What about the argument that low Fed Funds rates led to a scramble for yield among portfolio managers--and that yield was provided by Wall Street with leveraged instruments like CDO, CDS, leveraged bond funds, etc.? That is to say, low interest rates did not so much cause a mis-allocation of capital in the real world, but in the investment world through the overuse of leverage.

This is why the current Fed policy of keeping rates low but promising to raise them later is so counterproductive. Investment professionals are just smart enough to not be burned on the same stove twice, so they aren't willing (or are unable to convince their prime lenders to allow them) to employ the level of leverage that they used the last time around since they believe that at some point the Fed will begin to raise rates and squeeze the whole game again. Meanwhile the Fed is slowing economic activity in the country by refusing to allow savers to earn a safe return on their money in CD's, savings accounts, money markets, etc. To me, the solution is to immediately raise the Fed funds rate to something like 4% and proclaim victory--that way, the savers could start spending more and the leveragers wouldn't be afraid to leverage.

I know this might sound wacky, but what evidence is there that low interest rates cause inflation and expand economic activity is all cases? Maybe, just maybe, there is a difference between lowering interest rates from 7% to 5%, and lowering them from 4% to 0.25%, or that the outcome of an interest rate cut may depend on what happened before it occurred? It seems rather silly to me to assume that the system will respond in the same way to ZIRP the second or third or fourth time it is imposed as it did the first, analogous to a cortisone shot. By the third or fourth shot the effect is not nearly so magical as the first.

The house I bought in Stockton California in December 2002 now has a market value of half what I paid. Stick that on your graph.

Perhaps Canada has not had the same problems as the US due to the fact that you must prove your are employed and have a down payment before getting a mortgage. Also, you cannot walk away from your morgage without claiming bankruptcy.

A lot to respond to. Amazing what can happen when the power is out from the snow. In order:

(1) Anon--Both US and Canadian indices go up in the early '00s, but the US goes up at a much faster rate, especially starting in 2004.

(2) David Pearson claims the analysis is superficial. Fair. This wasn't meant to be the end-all explanation. Rather it was just meant to present a data point that doesn't fit comfortably with the monetary policy story. The second-order effects story is a much more convincing story, but notice that it isn't really a monetary policy story. It is a corporate returns story and a regulatory story. Monetary policy ceases to be the proximate cause of the crisis. Perhaps it still remains a but for cause, but there are several of those.

The Greenspan "put" story is also ultimately a regulatory story, not a monetary policy story. The only way that CW, Ameriquest, etc. were able to keep up origination volume after the refi boom was to adopt exotic products that were designed for frequent refinancings and to lower their underwriting standards. Permitting either of these was a regulatory failure, both in terms of financial institution safety-and-soundness, and in terms of consumer protection.


(3) John: There might well be another asset bubble that happened in Canada. But we'd still have to explain why it didn't happen in the Canadian mortgage market.

I disagree with your point about ARMs. When rates are as low as they were in 2001-2003, the only way they are going to go is up. In those circumstances, a FRM makes much more sense. Maybe you think you'll grab low rates for a couple of years with an ARM and then refi into a FRM when values go up, but I don't think that's the explanation of the growth of ARM share in this period. The swelling of ARMs was not from traditional ARMs (e.g., 7/1 ARMs), but from new hybrid ARMS and pay-option ARMs. The trick with these was the low, _fixed_ teaser period, often at a below market rate. I don't think Canada saw these instruments. The popularity of these instruments for mortgage banks was because they were perfect for an originate-to-distribute model. Hybrid ARMs are designed to be refinanced--another chance to pocket fees.

You're exactly right that the monetary policy story can only purport to explain the creation of a bubble, not the collapse. That's another of its deficiencies.

(4) But What DO I Know raises a less frequently told, but more compelling monetary policy story. It's a supply-side story, rather than demand-side (supply of mortgage capital, rather demand for mortgage capital).

AMC--There is such data for the US. You can get a lot of it from the Federal Reserve's Statistical Release Z.1, in the Levels Tables, probably L.217 or L. 218. Inside Mortgage Finance also has

Barb's comment reminds me of something I've been meaning to blog about: there's a general misunderstanding about the recourse status of US mortgages. In only a handful of states (AZ and ND are what come to mind) are most residential mortgages inherently nonrecourse. In other states, whether or not a mortgage is legally nonrecourse depends on a number of factors, including the form of the security interest (deed of trust or mortgage), on the type of foreclosure undertaken, and sometimes on whether it is purchase money or refinance money. All that said, the legal status of whether a mortgage is nonrecourse is only half the story. The other half of the story is that even if there is recourse on a mortgage, it is often very hard to collect a deficiency--state property exemptions, limitations on garnishment, and, of course, the ability to file for bankruptcy all make it very difficult to collect unsecured deficiency judgments. The point here is that the issue usually isn't technical legal recourse, but that unsecured debt is generally very hard to collect.

I don't know how easy it is to collect unsecured deficiencies in Canada, but I'm guessing it is not much easier than here.

So all in all, yes, of course there are lots of factors that contributed to the crisis, and monetary policy was not irrelevant, but the problems with the Greenspan Fed were less monetary policy mistakes than regulatory ones. These two pieces link together, but they are distinct, and blaming everything on monetary policy misses the point that the effects of monetary policy are shaped by lending regulation.

Easy money will seek out trouble, and if housing finance has lax regulation, that's where the money will go. The quesiton isn't why there was a bubble, but why there was a _housing_ bubble in the US (and a handful of other countries), but not in rather similar economies, like Canada's.

Can't forget the last link in the chain. Home sellers.. They were so addicted to the fast money and lower underwriting standards (home appraisals)that they were refusing mortgage loans that required stricter underwriting standards like FHA. I was looking during that time and had pre-approval for FHA but the majority of the homes selling at the time specifically would not entertain FHAs... had to be conventional or exotic. That created a wave of artificial increases in home values, in turn raising ad valorem taxes, in turn raising the value amount of the home that had to be insured. That $95K home you paid $120K for through a pay option in its second and third year ballooned the property tax and insurance bills that were paid out of pocket if it was an 80/20 pay option arm.... yikes! Also that wave was counted on by the taxing entities. They budgeted for it and now its not there anymore and we are seeing budget shortfalls for municipalities. (which by the way seem to be good investments especially when they get stimulus funds... I have a Muni thats paying 7% right now)

I think bankruptcy law experts are over their skis when they try to analyze macroeconomics.

I agree US housing finance mechanisms were a v. big factor. But one topic I am surprised no one mentioned is that in Canada home mortgage interest is not deductible.

I think Prof Levitan's post supports something I have been saying for years in comments which is we need, at every stage of the economy, higher down payments, at least 20% in the case of equity capital in businesses,includin banks, and at least 20% in the case of home ownership.
Also, I think t

I wouldn't for a second pretend to be expert in macro. But I've seen too few macroeconomists with a really good a grasp on how the economy actually works on the ground (and a handful of prominent ones who say amazingly ignorant things). It isn't just all as simple as following the Taylor Rule, Phillips curves, etc. The messy microeconomic plumbing, which is highly shaped by regulation, is hugely determinative of the impact of monetary policy. Macro can tell us whether we're likely to see inflation or an asset bubble, etc. It can't tell us where in the economy we'll see that occur, however.

The home mortgage interest deduction is an overblown issue. Yes, it distorts economic decisions and there is consensus that it is bad policy, even if politically sacrosanct. But it's worth recalling that the subsidy is particularly to wealthier Americans (larger mortgage means more deductible interest), and they may care less about the marginal dollar of savings as well as be more inclined to own for other reasons (including other tax reasons, like deduction for local property taxes and lack of capital gains tax on housing appreciation). Canadian homeownership levels are pretty close to the US, which puts the importance of the deduction in US housing patterns in doubt. Moreover, the deduction has been around for a while; it hardly caused this crisis.

MT's last comment (aside from the misspelling of my last name) gets to an important issue--should we be regulating leverage? John Genakopolous has a nice new paper on the topic on SSRN. I think there's a lot to think about on the question of leverage regulation, not least the technical problem of ensuring that appraisals aren't elevated.

Sorry for misspelling your name. I have a sight impairment that makes it hard to read on a computer.

You write, "Canadian homeownership levels are pretty close to the US, which puts the importance of the deduction in US housing patterns in doubt."

That is not necessarily true. You can have the same degree of asset ownership in the two markets but (x) at different price equilibria due to the presence of a subsidy and (y) with higher leverage ratios due to the fact that what is subsidized is debt.

At one time a 30 year fixed rate mortgage was an aberration in Canada. Almost everyone had short term mortgages - about 5 years - that had to be refinanced when the term was up. When I was first told about this by some Canadian friends I was shocked, and I wasn't even a housing lawyer yet. At that time - mid to late 708s - the 30 year mortgage was so ingrained in our culture that even a non homeowner such as I was at the time "knew" that the only way to borrow money to buy a house was through a fully amortizing mortgage.

Has the Canadian mortgage market changed since then?

Listen folks, lots of things caused the housing bubble. You had a perfect storm of government, consumers, and firms misbehaving or, at the least, missing or misreading the signals. You end up with a housing bubble.

However, the fact is that the housing bubble is only incidentally interesting in the bigger scheme of things - after all, the dotcom bubble (or biotech or name your favorite past bubble) was pretty darn big but didn't threaten the entire financial system. In my opinion, explanations for the housing crisis are pretty easy and, I suppose, moderately important, but don't get to the crux of the matter. Something else must have been going on to get us into the mess that we face.

To get at the bigger picture, think of things in this way. Housing, and subprime in particular, was the hand grenade. That hand grenade was tossed into a room full of TNT. The TNT was way in which financial firms increasingly fund (or funded) their operations through short-term repos (in which a big institutional investor gives a bank short-term, often overnight, funds in exchange for collateral). The grenade - housing, subprime or whatever - goes off, raising questions about the solvency of the bank or the value of the collateral, and you have a classic run on the banks - the institutional investor either won't renew its repos at all or requires a large haircut (i.e. discount) on the collateral. As a result, the bank doesn't have enough money to fund itself and must sell its assets in a fire sale. However, this run occurs at the wholesale level, rather than in the standard depositors-in-line way (largely due to deposit insurance). Most people don't actually see the run, except in cryptic newspaper headlines. Even when actual depositor bank runs took place, as in the case of Northern Rock in the UK or IndyMac in the US, those runs occurred after the damage from runs in wholesale funding.

Now, the maturity mismatch between assets and liabilities is a classic issue in banking. A bank has illiquid long-term assets (e.g. mortgage loans) that are funded by relatively short-term liabilities (e.g. deposits). (As an aside, securitization was largely an attempt to improve the liquidity of those long-term assets.) If everyone shows up at once to claim those liabilities due to some negative shock, the bank is in serious trouble - it doesn't have funds available to honor those redemption requests, rather they're tied up in long-term assets which would have to be sold, in a fire sale, to raise the required funds. This is nothing new - it's the way that banking, and banking panics, have always happened. (The Austrians would argue that this is an inherent feature of fractional-reserve banking. Of course, they're right. The problem is that their solution, while undoubtedly safer, would be horrifically inefficient during standard conditions, that is, when not facing a run. Alternatively, jack up capital requirements, but the Austrians have another point here - with fractional-reserve banking, you likely wouldn't have enough to withstand a run. Arguably, that's why we have a lender of last resort known as the Fed. But we also get moral hazard. Out the whazoo.)

The key to the current crisis is that the magnitude of both the maturity mismatch and the amount of assets being financed that way grew enormously over the last decade. Bear Stearns was funding a large part of its balance sheet through overnight repos (incidentally using a lot of securitized mortgage stuff as collateral). Which explains why a firm of its size could not only blow up, but could blow up so incredibly fast. In aggregate, I don't think that anyone knows the exact numbers, but we're talking about something in the ballpark of $12 trillion being financed through this type of short-term wholesale funding. Which explains why the whole financial system was at risk when the run occurred.

What's the point? The point is that all of this discussion - the Fed is to blame, Fannie and Freddie are to blame, a Consumer Financial Protection Agency is the solution - misses the point. All of that is targeted at the hand grenade. But the next hand grenade is just sitting out there waiting. And despite the best intentioned regulation that tries to target the hand grenade, you're not going to get it. There's simply too much money at stake. The Wall St firms will look to find the next hand grenade to mine it as far as they can. Alas, we've also shown them that we'll bail them out when the hand grenade hits the TNT.

And, at the end of the day, that TNT is still sitting out there - perhaps pulled back a bit, but it's still out there - just waiting for the next hand unforeseen grenade to light things.

Chabot--repos were the TNT for financial institutions, but there's still the question of what ignited the explosion, and that is the housing market. A CFPA wouldn't fix the repo issue, but it could ensure that consumer finance wasn't the source of problems. Moreover, even without the collapse of '08, we'd still have a lot of consumers really hurting in their mortgages. That's what the CFPA is aimed at preventing, not the transmission of firm collapse to market collapse.

But Adam, you're missing the point. Why waste all of this time, and political effort and capital, on an issue that is only tangential to the entire crisis? After all, you started this post with the claim that "(a) popular explanation of the financial crisis lays the blame at the feet of the Federal Reserve for lax monetary policy." In a literal sense, that claim is correct as would be the claim that eradication of subprime mortgages would have "prevented" the financial crisis. But we have precipitating causes and underlying causes. My point is that the precipitating causes are much less important, in the big scheme of things, than the underlying causes. Much of the policy discussion is correspondingly misdirected - you can fiddle around the edges all you like, but if you miss the TNT, you miss the entire issue. We fix the Fed, or fix lending practices, and we then sit back and pat ourselves on the back for a job well done. I think not.

One factor overlooked in the much greater tenant protections in most of Canada versus most of the United States. Tenants in British Columbia, which provides fewer protections than the large Eastern provinces, have rights that tenants in most of the United States can only dream of. Most tenants in the United States can receive a rent increase of any amount at any time, and can be evicted for any reason or no reason at all with a 30-days' (sometimes, less!) notice. So it's not surprising that tenants would decide that it's worth the risk of an ARM to get into a house, any house at all. Canadian tenants would be making a different calculation, and might well decide that their rentals are just fine.

The comments to this entry are closed.

Contributors

Current Guests

Like Us on Facebook

  • Like Us on Facebook

    By "Liking" us on Facebook, you will receive excerpts of our posts in your Facebook news feed. (If you change your mind, you can undo it later.) Note that this is different than "Liking" our Facebook page, although a "Like" in either place will get you Credit Slips post on your Facebook news feed.

Categories

Bankr-L

  • As a public service, the University of Illinois College of Law operates Bankr-L, an e-mail list on which bankruptcy professionals can exchange information. Bankr-L is administered by one of the Credit Slips bloggers, Professor Robert M. Lawless of the University of Illinois. Although Bankr-L is a free service, membership is limited only to persons with a professional connection to the bankruptcy field (e.g., lawyer, accountant, academic, judge). To request a subscription on Bankr-L, click here to visit the page for the list and then click on the link for "Subscribe." After completing the information there, please also send an e-mail to Professor Lawless ([email protected]) with a short description of your professional connection to bankruptcy. A link to a URL with a professional bio or other identifying information would be great.

OTHER STUFF