Narrow Market Exits Today Are Of the 2007-2008 Variety
The history of the John Hancock Tower in Boston is quite colorful, to say the least. Begun in the 1970's skyscraper boom, there were problems really from the start. In trying to lay the foundation, retaining walls meant to keep Back Bay's unstable soil from intruding weren't up to the task, creating damage throughout the local neighborhood. Even historic Trinity Church was impaired, as the transept part was nearly collapsed before the deficiency was discovered and repaired at some great expense for John Hancock. The building's windows were overly prone to detaching and falling to the pavement. In a windstorm on January 20, 1973, the still-under construction tower lost 65 panes, with each weighing some 500 pounds.
The loss of windows continued for some time. By April 1973, the unfinished building was, according to the Boston Globe, covered with more than an acre of plywood. Wherever missing glass panels were found, with any remnants littering the closed-off streets in the immediate vicinity, the holes were covered over with sheets of plywood. Apparently, clever Bostonians began referring to the structure as the Plywood Palace until all 10,344 panes were eventually replaced.
On a more serious note, structurally the building was never fully assured. It had terrible wind problems that caused more than a few cases of nausea in the upper floors whenever the wind accelerated. To provide more stability, two massive lead weights had to be installed on the 58th floor, at some considerable additional expense.
But in 1975, three years after the building was originally scheduled to open, an engineer from Switzerland told the Tower's owners that collapse was a very real possibility. The unique design was perhaps not fully understood for the area in which it was intended, as the parallelogram shape produced wind loads that, to say the least, were unanticipated. What the Swiss expert proposed was a very real wind scenario by which the building would fall over on its narrow edge.
Because of its long side face of 300 feet, the building was rather unbalanced, potentially, with not enough stiffening to brace against a wind load that might remove the structure's own internal balance or plumb. In other words, the long sides were plenty stiff but that didn't necessarily transfer over to the short side. In fact, what the owners were being told was that such was a unique weakness that threatened total catastrophe in realistic scenarios, not just worst cases.
It was there that the building finally encountered some luck. As it turned out, again according to the Boston Globe, there just happened to be enough room around the central service corridor to install 1,500 tons of diagonal steel braces to stiffen out any of these lateral wind loads along the long side.
Despite the existential threats and numerous repairs and delays, the building opened, finally, and took its place atop the Boston skyline as a really beautiful example of architectural advances. As a modernist, minimalist skyscraper it bridged the historical divide between other minimalist projects in the 1960's (like the former Twin Towers in NYC) and more recent introductions. I may be straying far afield here, but the John Hancock Tower really would not look too far out of place in any of the cities of the world built during the current skyscraper boom, despite its foundation being dug (badly) over four decades ago.
As such, the building took on the role of a "trophy" property in the Boston area, indeed within the whole United States. It was certainly a feather in the cap of John Hancock itself, though the winds of financial change opened up further episodes. By 1999, the stock bubble had blown into the insurance business and even the largest insurers taken under the structure of a "mutual", that is, owned by the policyholders themselves, were being derided as anachronistic and not fit for the age of shareholders. The "booming" stock market only made the impulse worse, as management in mutual after mutual began to commit to the process of "demutualization."
There was a fair amount of contention then about that structural shift, particularly in how John Hancock itself handled it in giving policyholders rights (few) and access (none) to the new IPO, but that is not my concern here. Not long after John Hancock "went public" in 1999, the stock market turned bear and management began seeking alternatives - these things usually top tick "markets."
By 2003, there were talks with other insurers and by 2004 John Hancock was being bought by Manulife of Canada. The merger proposal, as it was with synergy dictating some cost structures and the disposition of properties, suddenly opened the John Hancock Tower to the property "market." Beacon Capital Partners acquired the apex building in 2003, along with some other properties, at a total of $926.8 million. The loan-to-value on the deal was just a shade over two-thirds (with Beacon putting down about $304 million in equity).
Of course, the times being what they were, the building was sold only a few years later, this time for more than double the price. Nobody seemed too concerned when Broadway Partners, a New York hedge fund, laid down $3.4 billion for nine properties including the Hancock building. That valued the skyscraper and its garage at $1.35 billion, which was twice the deal value from 2003. At the time, that was just business-as-usual, particularly as so many hedge funds had charged into commercial real estate in this "wholesale" manner.
Sam Zell had just sold out 534 buildings to Blackstone Group for a reported $39 billion. The introduction of leveraged vehicles had seemed to rapidly conglomerate major property transactions, and so the field was growing condensed into the largest players with the deepest pockets; or, more precisely, the strongest connections to prime brokers and their massive "need" for production volume and yield.
It was even typical for bulge bracket "banks" to sponsor their own funds and then fund them through their own bank. Goldman Sachs had run and funded its own brand of property-focused funds going all the way back to 1991, including the very public purchase of Rockefeller Center in 1996. Known as Whitehall, the various funds became serious buyers in the 2000's, but it was 2007 and 2008 where they found themselves busiest. According to the Wall Street Journal, Whitehall funds of whatever individual vehicles had committed some $8 billion in those two years alone.
Goldman had tapped its own network of private clients, its own bankers and even the firm's own equity to fund each vehicle. And then it borrowed heavily to lever-up each fund into multi-billion dollar scales, often from Goldman the prime broker. They were especially interested in hotel properties, engaging in several LBO's and other "dark" transactions, while also targeting signature properties like NYC's Helmsley Building. Whitehall 2007 and a partner bought that apex property in December 2007 for $1.15 billion.
You know what comes next, the broadest, sharpest and deepest shakeout of the last eighty years. The Helmsley Building that Whitehall bought for over a billion was "worth" less than 75 cents on the dollar just a year later. The Whitehall 2007 fund's equity was marked at 50 cents, while Goldman struggled to hold on to other deals and vehicles without taking too many hits. In purchasing the Stratosphere Casino in Las Vegas in 2008, Whitehall had guaranteed $200 million of the $1.1 billion "mortgage" that Whitehall had borrowed from Goldman itself. That set up the strange process whereby Goldman Sachs would seize property from a fund in which it was already a significant owner.
Broadway Partners shared the same plight. In paying such a hefty price, part of that rationalization was that a signature property like the John Hancock Tower could realize a much higher rental structure. The top floors were rented at around $50 per square foot, but Beacon believed, in 2007, that they could eventually see $70. But in actually getting the deal done, Beacon had to take on several pieces of debt, including mezzanine pieces (akin to second mortgages) with higher rates and shorter maturities. The owner was facing $733 million mezzanine maturities coming due as soon as January 2009.
As the economy began to tank, renters were demanding decreases in rates not the increases that the various financial analysts had extrapolated. Property appraisals began to slump, which made refinancing that much more difficult in an already extreme environment. Then came panic, which effectively removed any ideas of refinancing at all.
In the meantime, several funds and financial vehicles that remained flexible enough had begun to see bargains amidst the chaos and carnage. Normandy Real Estate Partners (New Jersey based) and Five Mile Capital (Connecticut) began buying pieces of mezzanine debt at fire-sale prices all over. Among them was debt tied to the Hancock Tower in Boston, which the two funds acquired from none other Lehman Brothers (desperate itself for cash after being a significant provider of property-based leverage) and RBS for a reported 30 cents on the dollar. All they needed was a default event and their position in the capital structure would allow them significant advantages.
That happened in March 2009, as Broadway finally defaulted on secondary loans, and an auction took place that saw only two funds make bids - Normandy and Five Mile. The stated value of the transaction for the John Hancock and its garage was a rather shocking $660 million, or less than half of what was paid just two and a half years before. That, however, doesn't actually figure the full-scale of what transpired, and just how low valuations had sunk.
Part of the "value" in taking the tower meant assuming some of the mortgage from the prior owner at very favorable terms. When only two bidders show up to a default auction, terms tend to be quite generous in that direction, especially when those bidders already acquired a big seat at the table. What essentially transpired was that Normandy and Five Mile "bought" the John Hancock Tower for $660 million, negotiated write-downs of parts of the debt, for no money down whatsoever apart from their pennies on the dollar purchases of smaller debt pieces prior to auction. But the value of acquiring such favorable financing was something like $200 million itself, so the effective purchase price was really less than $500 million for a property once believed "worth" close to three times that.
It was a hell of a gamble and it paid off large in late 2010. Another firm, Boston Properties, bought the John Hancock Tower and its garage for about $930 million. The new owner assumed the $640 million mortgage and paid in $290 million in equity. For Normandy and Five Mile, they made leverage, and illiquidity, work in their favor to a massive degree.
In some ways that is as it should be. There needs to be nimble (and smart) firms and individuals that have kept their heads and are willing to take huge risks to essentially provide, eventually, a floor to "market" prices. That is what prices are supposed to do, unfiltered and unrestrained they will eventually clear out imbalances and get the entire affair restarted. In one sense, however, there is clearly a problem with prices being determined by transactions rather than any sense of "worth" or "value."
That is an ages-old problem, as it was "settled" long ago that something is worth what somebody will pay for it - and there is a lot of weight to that sentiment. However, what is lacking is that there is often no appreciation for what it actually takes to maintain order in just such a design. In this case of financial transactions for opaque and hard-to-value property, liquidity gains paramount placement in factoring systemic function.
This is essentially the idea of how big are the exits in times of less-than-ideal circumstances. In pure monetary terms, like that of the Great Depression, that was a money stock issue as the massive rise in the desire for currency was never going to be met by a banking system that had fractionalized far too much of it. The issuance and maintenance of currency liquidity was never rescaled in the 1920's to the size of claims for it. Thus, when a huge proportion of deposit holders rushed for the exit at once, they were far too narrow to accommodate and the whole system collapsed on itself as depositors were forced into drastic (and truly deflationary) measures to gain just simple access to cash. Prices were altered drastically lower, resetting systemically the entire global economy.
In the shadow banking system as it looks now, or even since 1990, money stock is far less important than other factors like equity capital and flow. Liquidity takes on that additional element whereby it doesn't necessarily matter "how much" there is available, but rather how it gets from A to B (and further to C and D and beyond). The mechanisms in this masterful fit of mathematics are often opaque and esoteric places that few want to venture into. That includes repo markets where, in the light of the 2007 collapsed state of unsecured interbank lending (http://www.realclearmarkets.com/articles/2014/08/08/anniversary_of_a_panic_that_changed_everything_101216.html), nearly all marginal liquidity was maintained and offered.
It was from that point at which the bottleneck appeared and corroded. Uncertainty about collateral meant uncertainty about funding, which meant banks that usually filled prime brokerage roles to offer persistent and continuing leverage could or would not. Balance sheet capacity, which is the measure of true liquidity, shrunk so heavily that the scale of indebtedness built up since the 1990's could not be maintained - the exits were far narrower than anyone anticipated, especially given the scale of leverage and debt employed in taking asset prices as high as they were. So much credit and debt was dedicated to price appreciation that the idea of liquidity was never matched to the potential need for maintaining its order. Most of that was a fit of recency bias, as nobody believed that liquidity would become so impaired simply because it never had under the "modern" operation.
We think such problems in the past, left behind with the rebirthed financial system that has been declared "healed" for several years now. But there is a problem with that view that has come into focus again recently. The repo market underwent some severe strain in June, though subsiding by July. However, that stain has again appeared in the form of another surge in repo fails.
There are several implications of this, from the technical all the way up to more systemic cracks. It is there that I think this warrants much further attention, as what this repo strain tells me is of a similar mismatch between liquidity, or what we think might be available, and the scale of credit and debt offered that is now sustaining prices (though in other areas than property this time). One more time it appears as if the exits are not just far too narrow, they have actually become smaller and frailer in the past two years at exactly the same moment the pile of debt and credit has attained levels far surpassing even that of 2007.
A good part of that problem has a heavy regulatory flavor as credit dealers clearly have undergone a profit transformation as it relates to considerations of balance sheet capacity. In the age of Dodd-Frank and Basel III, dealer inventories have shrunk precipitously as the cost of maintaining collateral stores and acting as a potential liquidity provider has been enlarged. However, part of that cost, as the repo market in the past few months suggests, is not simply regulatory alone.
For some reason dealers are holding far less US Treasuries, which is not inconsistent on its own with these regulatory cost burdens. But the timing of this erosion, and indeed a clear diminishment in the volume of repo financing itself, dates to just October 2013 (http://www.alhambrapartners.com/2014/08/20/searching-for-fail-and-still-finding-it/). That is right around the time of maximum pain in all parts of the credit markets, such as interest rate swaps, curve trades, and plain bond prices. Then-Chairman Bernanke's taper threats of May 2013 had set in motion a massive global dollar disruption that by October had rippled into the banking system - but it was odd that it ever accrued such massive reach.
Banks had already begun to layoff huge proportions of their mortgage staff by October, in anticipation of what was judged to be a far less profitable mortgage market from that point forward. The reason for the mortgage disruption was more to do with MBS trading and price behavior than any reasonable judge of carry spreads. In other words, banks were making more money on price movements in MBS, under the implicit cover of QE3 that they believed semi-permanent, than on the traditional factors of mortgage lending.
I think the same carried over in the treasury market, particularly as treasury rates were blowing much higher by October 2013, taking the treasury curve much steeper. That was a very bad place to be if you were holding a huge inventory of UST and attended that with an immense predilection to receiving fixed premiums in interest rate swaps. In short, balance sheet capacity of the kind that would "expand" the exits was no longer so profitable under the threat of taper, and its final realization a few months later.
The results, I believe, show exactly that. Credit market dealer capacity has eroded steadily and significantly since then - even further than the diminished capacity already apparent due to the regulatory changes noted above. The net effect of all of this is as it was in 2007 and 2008 - narrow exits that are not suited for the scale of claims on them. The extreme amount of debt at this point, despite all the public proclamations of deleveraging and whatnot, is very much disproportionate to the potential pricing problems should conditions once more trend toward less artificial order.
Liquidity is not what you see today, it is what is available when you need it most. That is the problem, because it has a broad tendency to disappear under stress. That is why you would very much like to see liquidity operate without fail or falter in the best of times, so that you might hope for something decent in the worst of times. The huge declines in dealer volumes of corporate credit, the massive and incongruent MBS selloff of last year and now the repo fails in UST are not favorable signs for liquidity under relatively benign conditions.
