Getting Back on the Horse

It’s an old saw that, when learning to ride, or swim, or master any inherently risky activity: when the horse throws you off, the important thing is to get back on the horse. It’s important to understand what happened to you, of course; but, even more than that, rote confidence is required.

In that context, is the world ready for the securitization of cancer research?

Securitization, you’ll recall, is the bundle of techniques that received much of the blame for the recent crisis. Securitization involves taking a slew of ordinary debt contracts – not just mortgages, but car loans, credit card receivables, student loans, whatever – directing their cash flows into one giant pool, and then separating its  streams, structuring them in tranches in accordance with differing preferences for return and risk, and then selling off, as bonds, these reassembled packages of repayment promises to investors, eager to earn some interest on their idle cash.

The word itself, a neologism, didn’t meet The Wall Street Journal’s standard for editorial clarity when Salomon Brothers employed it to describe the first conventional mortgage pass-through security in 1977.  Fifteen years later, securitization was being trumpeted as one of the major financial innovations of the age. And fifteen years after that, “originate to distribute” was being blamed for the panic of ’08.

Here’s how Bethany McLean and Joe Nocera put it in All the Devils Are Here: The Hidden History of the Financial Crisis:

Before securitization, lenders had to care about the creditworthiness of borrowers.  They held the loans [they made] on their books, and if a barrower defaulted they took the hit.  That’s why borrowers who didn’t have much money couldn’t get mortgages: lenders were afraid they would default. Securitization severed that critical link between borrower and lender. Once a lender sold a mortgage to Wall Street, repayment became somebody else’s problem. The potential consequences of this shift were profound: sound loans are the heart of a sound banking system. Unsound loans are the surest route to disaster. But at the time, almost no one seemed to realize that the wave of poorly underwritten loans that securitization seemed to encourage was a monstrous red flag.

There is no doubt that widespread uncertainty about who owed what to whom under what circumstances was a key part of the problem.  Indeed, problems with securitized products go back at least as far as farm railroad bonds in the Panic of 1857. And the mortgage-related credit craze and housing boom, with its assumption that home prices would never decline, surely deserves a special place of its own in the Bubble Hall of Fame. What caused the white knuckles, though, and cost so much money, was halting the run, the short-lived period of incipient panic, when nearly everyone thought of taking their money, not just out of the bank, but out of the money-market fund, the securities markets, the private financial system altogether.  For many market participants, the world had become suddenly, briefly, unfathomably complex.

It turns out that most asset-backed instruments, including derivatives, held their value quite well through the crisis.  Bondholders, those whose problem repayment had become, for the most part did fine. Securitization thus proved to be a powerful and mostly safe way of raising capital and sharing out risk in the modern world, more efficient than traditional institutions.  That’s why the practice was so widely adopted in one industry after another – anywhere a shortage of funds, an abundance of investors and an adequacy of spreads was sufficient to make a market. In an age of computers and cheap communication, the techniques of securitization are no more likely to be banished than other major inventions of the past fifty years, including integrated circuits, packet switching, or search advertising.

Now Andrew Lo, of the Massachusetts Institute of Technology, wants to adapt securitization again, this time to pay for more and better cancer research.  In an issue of Nature Biotechnologyearlier this month, Lo and two of his former students, Jose-Maria Fernandez, of MIT’s Sloan School and its Laboratory for Financial Engineering., and Roger Stein, of Moody’s Corp., published Commercializing biomedical research through securitization techniques. Lo, it should be noted, is not just another hopeful young man with a bright idea. A successful hedge fund operator (chief investment officer of AlphaSimplex Group), he was a frequent consultant to the US Treasury Department’s new Office of Financial Research and has been a leading contender to head it. Here is an early shrewd take on the plan, by Felix Salmon, of Reuters.

Almost any income stream, actual or potential, can be securitized.  Music fans will remember Bowie bonds. They were the creation of high-flying structured securities salesman David Pullman (Ziggy Gold Dust, The New York Observer dubbed him at the height of the dot.com craze). The idea was that royalties from the sale of David Bowie’s albums were so stable that Prudential Insurance would give the rock star $55 million in exchange for a roughly ten-year claim on his earnings, at a rate 2.6 percentage points more than US Treasuries were paying. When Goldman Sachs tried the same thing with Bob Dylan’s songs last summer, the offering met resistance, and Goldman pulled it from the calendar.  Nor are such deals unknown in biotech. Royalty Pharma, HealthCare Royalty Partners and DRI Capital all have royalty monetization businesses.

The difference is that Lo and his co-authors want to do it prospectively, raising a fund to support a large fund – a megafund, they call it – to invest in a carefully diversified portfolio of around 150 little research ventures in university-affiliated laboratories around the country. In principle their scheme is not much different from those buccaneering companies that go searching for various undersea shipwrecks. In fact, the scheme rests on a careful simulation using historical data for new molecular entities in oncology that appeared between 1990 and 2011.

If the success rate in the future were pretty much like that of the past, they say, megafunds of $5 billion to $15 billion could yield investment returns of 8.9 percent to 11.4 percent to equity holders and 5 percent to 8 percent to holders of research-backed obligations – returns too low for venture capitalists, but just the thing to interest pension funds, insurance companies and other big institutional investors.

The opportunity exists because of the widespread consensus that the current basis for developing cancer drugs is not working, despite the vast sums being spent.  Profit-driven investing in giant pharmaceutical and biotech is providing few dramatic advances; private equity ventures don’t fare much better, thanks to the vagaries of the business cycle and the so-called “valley of death” – the gulf between the kind of basic research that government supports through grants to teams of scientists, and the expensive clinical trials that companies must undertake before they can bring new drugs to market. In 2010, for instance $40 billion was spent on basic research and $125 billion on clinical development, but only $6 billion or $7 billion on what is called “translational” research to bridge the gap.

Suppose that such a megafund raised, say, $15 billion.  To spend it wisely on a “preclinical incubator” would require talent of a high degree. A blue-chip science board would have to be assembled; the most promising projects identified; trustworthy underwriters hired; a management team empowered to quickly cut funding to the losers and briskly push promising projects forward.  Megafund managers would behave more like traditional venture capitalists, immersed in the science and engineering aspects of their projects, Lo says, whereas biopharma firm executives who oversee one or two enormous projects have little choice but to act as personnel managers. At every juncture, the megafund idea would be to skew incentives in favor of ultimate success. Instead of a war on cancer, as the slogan has it, why not put a price on its head?

And suppose the first megafund were successful?  Wouldn’t a second megafund be raised, and a third?  It wouldn’t be the first time that securitization worked too well. Raising money is easier than spending it wisely; spending it, wisely or not, is easier than generating the cash flow necessary to service megafund debt. And sustaining the kind of political and regulatory oversight for such mechanisms to flourish may be the most challenging task of all.  And yet the promise of such mechanism design seems very great.

Lo and his co-authors have located their proposal in the context of the growing trends towards “venture philanthropy,” as practiced by organizations such as the Gates Foundation and the Robin Hood Foundation.  Such charitable giving is highly valuable, they say, as far as it goes; so are various programs of government sponsorship of translational work:  but none go far enough. The pool of investment capital seeking a reasonable rate of return is enormous. Proposing to raise billions of dollars through securitization to fund biomedical research opportunities that go begging under the current system may seem like a pipe dream in the current political atmosphere. (“Securitization is clever but not magic,” says a friend who tracks  portfolio performance for a living: “How much would you put into securitization for transporter beams?”)  But in the province of experts, it just might (slowly) work. MIT is planning a major conference around the idea next year.  Lo’s plan is an act of leadership.

The larger issue here has to do with getting back on the horse.  There are other, more important proposals in the offing as regards securitization – in particular the idea of interposing a layer of federally chartered and closely supervised banks between the originators of securitized debt and its ultimate beneficiaries.  Such “narrow-funded banks” would constitute the sole market for asset-backed securities (they could buy government and agency bonds as well); they would finance their purchases by issuing medium- and long-term debt and through repurchase agreements.  They would make their money on the spread; competition among them for the relatively easy pickings would do the rest.  The effect would be to create a dull and only moderately profitable industry whose sole purpose would be to put money where heretofore only the credit rating agencies’ mouths had been.

The key concept here, at least for everyday political discourse, is rote, meaning by automatically,, without thought or even any very deep understanding on the part of most people: the way you slip bach into flying after news of a gruesome crash.  Securitization is not like nuclear power or supersonic transport, a technology abandoned as intrinsically too cumbersome and/or dangerous before barely half a century has passed. Financial engineering is here to stay.  Until the everyday safety of its techniques once again become ordinarily taken for granted, like the engineering of tall buildings, airplanes, computer systems, the economic recovery will hobble along at a lower rate than before.

Enlace