August 2004

Different Bubble, Different Outcome?

The Inverse Of Multiplication...In case you've forgotten grade school math, that's division.  Well, there's certainly no question that there's one thing that has multiplied like wildfire over the last half decade, and that's residential real estate prices.  As you know, the debate has currently appeared in the mainstream as to whether a bubble exists in household real estate asset values.  A few weeks back, in a study published by the clairvoyant folks at the Fed, the emphatic declaration was "no way".  (Remember, the Fedsters are currently 0 and 1 in terms of "a priori" bubble recognition.  Luckily for them they still have a chance to redeem themselves if they get to work now by correctly identifying any one of the current number of other bubblicious asset classes of the moment.)  In response to the documented Fed real estate appraisal, a research report put out by HSBC officially and professionally concluded "yes way", bubble conditions do exist in the land of residential real estate prices.  We're not going to debate the bubble characterization issue except to say that in our minds, bubble conditions exist in real estate finance.  Although we have some pretty solid guesses, how the ultimate ebb and flow of the mortgage finance cycle plays itself out remains to be seen.  

You'll remember from discussions past just how meaningful residential real estate as an asset class is to US households at the moment.  Here are some facts from the recent Fed Flow of Funds report that documents the story and pretty much puts asset class exposure into direct perspective, especially over the last half decade.  (To be fair, we've included household mutual fund holdings in the common stock asset class category along with individual stock holdings.)

Clearly, the first and most obvious observation is that accelerating residential real estate values kept household net worth largely moving forward during one of the greatest equity bear markets in many decades.  As of 1Q 2004, household net worth rested at an all time high.  And so did the value of household real estate assets.  As you can see, despite the recovery in equity markets, household exposure to common stock is below where it was at year end 1998.  Secondly, it's clear that real estate has become a much more meaningful asset class to households relative to their common stock holdings over the last four years.  At year end 1998, household net worth exposure to common stocks and residential real estate was about equal in dollar magnitude.  As of the end of 1Q, household residential real estate values exceeded common stock holdings by 67%.  In our minds, this has put real estate front and center in terms of the household "wealth effect". As mentioned, we've been through this conceptual material before.  Suffice it to say that residential real estate values are very meaningful to the total current picture of US household net worth, to say nothing of emotional and financial well being and confidence.

Different Bubble,  Different Outcome?...Without us having to say it, it's a pretty good bet that a downturn in residential real estate values would not be good for forward household confidence.  But if a real estate downturn were to become serious enough to trigger meaningful financial defaults, the potential consequences become much more dark than just households feeling blue.  Much more dark for the entirety of systemic credit expansion stateside, and the Fed's theoretical ability to revive and/or stimulate the broader financial and economic system.  Essentially, we're referring to the popping of a price bubble.  What we're leading up to here is that the very fact that asset bubbles, or potential assets bubbles, exist does not necessarily pre-determine massive fallout across the broader economy in the event of their popping, but rather it's how they are financed that just might be the key issue in terms of potential economic and financial fallout.  As you know, at least so far, the popping of the NASDAQ bubble four years back did not take the US economy directly into some type of deep recession or semi-depression.  At least not in terms of headline economic stats.  Certainly a lot of folks lost a lot of real money, but one of the keys to economic and financial survival so far has been the fact that lending institutions of all types were about to embark on a massive interest driven lending spree.  Broad potential for credit expansion was not hampered in general by the NASDAQ price pop.  And it's really no secret as to why.  Quite simplistically, the stock market bubble was not financed by the banks.  It wasn't financed by Fannie and Freddie.  At least not directly.  It was financed by pension funds, mom and pop investors armed with personal, IRA and 401(k) money, and it was financed late in the game by foreign interests unable to avoid the temptation of joining the party.  And, yes, in part it was financed by margin debt.  But compared to mortgage debt in the system, margin debt even at its peak was an absolute dollar rounding error.

Real estate is different.  At least as far as the banks are concerned.  And in a severe real estate downturn, FNM and FRE would probably be basket cases.  At best maybe it's just their equity holders that would handle basket case duty.  The big difference in the popping of the equity bubble versus the popping a potential residential real estate bubble at the moment is the financing.  As we have shown you before, the US banking system is a huge player.  Remember, it's not just Fannie and Freddie financing every piece of real estate in this country.  The numbers simply speak for themselves.  As you can see, the commercial banking system in the US is pushing toward $2.5 trillion in total loan exposure to real estate, both commercial and residential.  Also critical to keep in mind is that what you see below is only lending activities.  On the investment side of the equation, US commercial banks have plenty of real estate related "paper" exposure in the form of investments in agency debt, CMO's. GMNA's, etc.
       

And, in our minds, the big difference for the US financial system if real estate values were to "pop", so to speak, is to be found in the multiplier effect inherent in fractional reserve banking.  For the sake of conceptual argument, we'll keep the numbers really simple (for ourselves more than anyone else).  Let's use a bank deposit "reserve requirement" of 10%.  The bottom line is that every dollar deposited into the banking system can ultimately potentially "create" $10 of credit expansion.  With a $1 deposit, a bank must keep 10 cents in reserves and can then lend out 90 cents.  Assuming that 90 cents purchases an asset and the seller of the asset deposits the 90 cents into the banking system, 81 cents can then be lent out (90 cent deposit less the 10%, or 9 cents, reserve requirement).  On the very first lending transaction, we've now created $1.71 of new credit for the $1 dollar originally deposited into the system.  Extend the lending possibility math and $1 has the potential to create $10 in new credit.  Simplistically, this is the very mechanism by which we conduct modern day "fractional reserve banking".  No big mysteries.  It's a wonderful life, right?

But what happens when defaults occur in the banking system?  Well, the machinery is thrown into reverse, whether voluntarily or not.  When a bank loses a dollar through a loan default, its original deposit liability does not go away.  At the start of a default process, the bank simply loses $1 of its own equity.  It has to take money out of shareholders pockets to repay depositors if it loses their money in lending activities.  But, moreover, and a bit conceptually, the $1 a bank loses in a loan loss is another $1 it cannot turn into $10 of new credit expansion.  This is the double edged sword of fractional reserve banking.  Very simplistically, loan losses can beget credit contraction, dependent of course on the severity of system wide loan loss experience in any asset class.  Again, in our minds, a potential popping of the theoretical (for now) bubble in residential real estate prices could foster quite a different outcome for the real economy and financial markets than did the bursting of the NASDAQ bubble four short years ago.  The popping of the NASDAQ bubble took mom and pop money down the drain, it blew a hole in many a corporate pension fund, and postponed the retirement dates for many an IRA dependent household in the US.  But what the popping NASDAQ did not do was impede systemic credit creation.  Alternatively, a severe downturn in real estate values that triggered real mortgage loan defaults would, in part, act to set into reverse the US banking system fractional reserve multiplier mechanism.  A severe or prolonged enough downturn in real estate would, from our point of view, at the very least call into question the rate of change in US financial system credit expansion possibilities.  In other words, different bubble, very different outcome for the real US economy.  

As you know, we've experienced real estate downturns in prior cycles.  The late 1980's/early 1990's was the latest in what has been a series of historical replays.  And during these periods of historical real estate price downturns, banks have backed off in terms of their real estate related lending activities.  You'll see this in the chart below.  But what is also clear as a bell is that as a percentage of total bank loans and leases outstanding, US banking system exposure to real estate has never been higher than we now experience.  One quick note, what you see below does not include banking system exposure to home equity lines of credit.  That probably pushes the current number to something near 55+%.
       

Greenspan and the Fed can do all of the tough talking they'd like about standing ready to raise interest rates if inflationary pressures even sneeze.  But the reality is that they will not be able to tolerate a pop in the mortgage finance bubble.  That will not be acceptable as the fallout would seem much more severe systemically than was the case with the equity bubble burst.  And, as you know, we have not even mentioned potential impacts on the large GSE's that are holding a good chunk of the remain mortgage debt in this country.  Or the fallout a significant GSE problem would transmit throughout the system.  You remember the GSE's.  Folks like Freddie Mac, who still can't seem to be able to produce accurate financial statements and supposedly won't be able to for some time.  Just be patient, right?  Or Fannie who clocked in at 2.2% equity to total capital as of 12/31/03.  

What Do We Do For An Encore?...A few last pictures of life in the residential real estate world as we have known it over the recent past.  Unless mortgage interest rates plummet at least 100+ basis points or more from where they stand today, it's a good bet that the momentum of the recent refi cycle is over.  Is what you see below perhaps the ultimate head and shoulders chart formation?  For the sake of credit expansion fostered by the US banking system and GSE complex, let's hope not.
      

In recent months, we have likewise experienced record existing and new home sales, as well as record corresponding construction activity.  But as with refi activity, is mortgage financing beginning to peak on a rate of change basis?  Remember, what's important for the economy is not necessarily residential real estate prices, but rather the volume of financing activity occurring in the system.  That's what makes aggregate credit expansion go.  That's what has kept liquidity flowing into the real economy.
         

For now, the US economy and financial system has proven that it can withstand an equity bubble implosion.  Of course the price for that resilience has been record monetary and fiscal stimulus, record credit expansion, record trade and budget deficits, etc.  We're not so sure that a meaningful setback in US real estate prices would produce a similar outcome.  Especially since the monetary and fiscal authorities have largely plundered their financial ammo supply.  And especially given the fact that the provocateurs of recent systemic credit largesse, the banks and the GSE's, would take a direct hit to the balance sheet.  In our minds, all bubbles and not created equally nor do they deflate in similar trajectory.  For now, the system has been able to withstand the bursting of one financial bubble.  Let's just hope it isn't called on for an encore performance.

Taking Stock...Again, at the moment it's just a bit too early to call for significant mortgage credit defaults ahead.  Something like that just doesn't happen over night.  But we suggest that at the moment, the probability for this type of occurrence to unfold has not been higher in many a moon.  Roughly 35% of recent mortgage refi and purchase related activity has been ARM vehicles.  Total ARM debt outstanding in the US system right now is pushing 18-20% of total mortgages outstanding.  It's a very good bet that a once in a generation interest rate cycle has seen its best days.  For some time now, we have been harping on the fact that wages will be critical to the forward movement of housing prices as anomalies in financing begin to deteriorate.  And at least for now, wage growth in the US is negative in real terms.  The year over year change in domestic wages is running close to 1% below the year over year change in the already lowballed CPI.  Overlay on this the fact that 47% of total US Treasury holdings are in the hands of the foreign community, and really never has the potential for forward interest rate volatility in US financial markets been more of a question mark.  We're not suggesting that the end of the world lies around the corner.  You can see in the chart above that a little less than 1.3% of US mortgages are in foreclosure at the moment.  But we suggest that it won't take much in the way of defaults to spark a problem, whether real or emotional in the eyes of lenders.  As we mentioned, Fannie is sitting on just 2+% equity capital.  In our minds, key to the US residential mortgage credit quality equation ahead will be interest rate volatility and real US wage growth.  Simple enough?  A deterioration in mortgage credit quality will be a process, not an event.

For now, we need to listen to what the markets are suggesting.  After all, if a mortgage credit problem is to arise stateside, the markets will have at least begun to discount it well before the reality becomes a consensus viewpoint.  And at the moment, the market appears to be telling us that broader housing momentum is slowing.  This is where any longer term problem of heightened financial risk is going to start.  As you can see below, the Philly housing index (HGX) stands at a critical technical crossroads.  The price as of month end rests at the current meeting place of the 50 and 200 day moving averages.  And it is crystal clear that the 50 and 200 day MA's have not met up like this since the equity market rally began in early 2003.
      

Even more definitive from a technical perspective is the current picture of housing as described by the Dow Jones Home Construction Index.  The 50 day MA crossed through the 200 day MA to the downside last month and it sure appears that a pretty classic head and shoulders pattern is in place.  A break of the H&S neckline to the downside will be anything but good news for the broader housing industry.
          

If and/or as macro housing industry fundamentals begin to deteriorate, watching the mortgage lenders will become a necessary exercise along with monitoring mortgage delinquency data.  Specifically, we'll have our eyes on the banks and Fannie.  For now, the banks recently made a new high (the Philly Banking Index) and have returned to test the breakout.  Certainly one thing to be aware of when looking at this index is the influence of M&A in the index price at any point in time.  Surely the recent Fleet and Bank One acquisitions gave the BKX a little price boost that pushed the index into record territory.  Absent any other large take over activity ahead, we should know in relatively short order whether the recent break out to new highs was simply a fake out.
       

Maybe more important as the potential canary in the coal mine is Fannie.  What has struck us for a good while now is that despite absolutely record breaking refi and new mortgage activity during 2002 and 2003 (and into early 2004), Fannie has not been able to make a new high as a stock.  In fact after peaking in late 2000, Fannie is essentially putting in a series of relatively well defined lower highs.  We suggest that the ultimate resolution of the longer term wedge formation that is clearly obvious in the weekly chart below will be very telling as to mortgage credit quality stateside going forward.
       

Although it may sound a bit melodramatic, an even semi-meaningful problem with mortgage credits in the US financial system ahead will spell a very different outcome for the real world financial markets and economy than was the case with the popping of the equity bubble a few short years ago.  For ourselves, we'll be watching housing industry fundamentals, mortgage paper credit spreads, delinquency characteristics, and the stocks of those folks near and dear to the mortgage finance industry like a hawk. 

YEAR 2003 MONTHLY ARCHIVES

YEAR 2002 MONTHLY ARCHIVES

YEAR 2001 MONTHLY ARCHIVES

YEAR 2000  WEEKLY ARCHIVES