Francie Stevens: “I’ve never caught a jewel thief before! It’s stimulating! It’s like... It’s like...”
John Robie: “Like sitting in a hot tub?”
— To Catch A Thief
Part of the real pleasure of the interwebs for me, O Dearly Beloved, is the occasional opportunity Your Muddle-Headed Correspondent has to engage in a stimulating conversation with one or more individuals who are clearly more intelligent, better read, and cleverer arguers than Yours Truly. I tend to gain a lot from such interactions, if only a better understanding of my own opinions and the limitations of my knowledge and intelligence. They tend to generate a crush of sensations, including excitement, the sheer terror of discovery that I am out of my depth, and an urgent desire to land a couple of cheap shots on the champion before I scramble out of the ring to safety.
Today’s reflections are inspired by the response Carolyn Sissoko made to my recent post on unlimited liability in finance. While I am not equipped to address some of her points—especially on the topic of what she sees as the increasingly pernicious use of collateral by modern financial intermediaries—I would like to venture a few remarks and intuitions, in the spirit of a novice fighter taking a couple tentative jabs at Mike Tyson. Hopefully she will be gracious enough to let me run away thereafter.
First, Ms Sissoko uses historical evidence (what’s that?) to demonstrate that a regime of unlimited liability is not necessarily inconsistent with low required rates of return on capital:
The system of unlimited liability banking grew up in an environment with usury laws, so interest rates (on short-term debt) did not exceed 5% per annum. Market rates often fell as low as 2%. It’s far from clear that low interest rates for borrowers are inconsistent with unlimited liability on the part of lenders who choose to use their ability to borrow to leverage their returns (i.e. to act as partial reserve banks).But I think this point misses the thrust of my previous argument. My intuition about unlimited liability is that capital providers subject to it have a natural limit to the amount of credit they are willing to extend regardless of price. Each lender sets her individual limit based on her estimation of the likelihood of loss beyond initial capital invested. Beyond that there is no price at which she would be willing to lend. This certainly would be my preference: if I faced the loss of my home and possessions, penury, and utter ruination, you can damn well be sure I would not extend just one more loan to capture an extra fifty, hundred, or even thousand basis points of yield. I do not think I am alone among heartless, flinty-eyed rentiers in this regard. The historical evidence Andrew Haldane cites from early 19th century Britain is entirely consistent with this: compared to the situation today, banks were massively overequitized and highly liquid, and bank assets and hence lending were a very low proportion of the economy. If my intuition is correct, it may well be that prevailing market rates of interest during that period evidence less that capital was plentiful and demand fully satisfied and more that supply and demand were balanced in a regime of artificially limited supply. Certainly Mr. Haldane contends—based upon what, I do not know—that the system of unlimited liability was not capable of supplying the growing need for capital during the rapid industrialization of the mid 19th century.
I strongly suspect that if we attempted to reimpose a regime of unlimited liability on capital providers nowadays, the pool of capital available to the economy would shrink, and likely dramatically. People with capital would simply become unwilling to lend to or invest in risky projects beyond a certain point, and the evidence from 19th century Western industrial societies seems to indicate that point would be at a dramatically lower level than we have become used to over the past 70 years. Unlimited liability could well be massively contractionary. This, I assume we all can agree, would not be a good thing.
Second, Ms Sissoko rather loses me when she asserts that
from a theoretic point of view, a banking system doesn’t need capital, it needs trust (aka credit). If the institutional framework is carefully structured (that is, debts are enforceable, outright fraud is disincentivized/rare, etc.) there is no shortage of capital — capital is created out of thin air by a plethora of unsecured, but trustworthy, promises. Effectively, capital is cheap, because the institutional structure of finance reduces the risk of losses to a minimum.
It may be that the intestinal parasite (me) within the dinosaur simply cannot conceive of a world without dinosaurs or dinosaur intestines, but I don’t get this. Capital is, in my understanding, at base a buffer against loss. Perhaps capital and bank lending vanishes when you sum all of its offsetting accounts across the entire economy, but capital serves a very important protective function at the “local” level of real creditors and borrowers. Capital absorbs loss, and either stops or slows the propagation of that loss through the daisy chain of economic interrelationships economic actors maintain. It is like the collapsible drums filled with water at the top of a freeway exit, whose function is to slow or impede the destructive progress of a runaway car. Sure, a runaway car will eventually stop of its own accord due to friction and gravity—if not another car or building—but do we really want to conclude from this that crash drums aren’t desireable or necessary?
I presume Ms Sissoko would counter that, with appropriate care and attention, we could design freeways and perhaps even cars so that we need not worry about collisions or runaways, but I fear that implies a level of omniscience—and systemic rigidity—which I find hard to credit. Financial loss is triggered both endogenously and exogenously. Being unable to anticipate all the ways financial loss can occur and directions from which it can come, I would much prefer to have various pools of capital sitting around the system, hopefully helpfully positioned between me and disaster. Besides, without capital, how can we enforce incentives? Capital belongs to someone. I thought the point was to reduce unnecessary systemic risk by presenting those positioned to create the risk of loss in the first place with incentives not to do so recklessly. How else can we do this except by making them the first to bear that loss; i.e., lose their capital?
One of the most pernicious effects, in my opinion, of the evolution of limited liability in the financial system, and the consequent transfer of more and more tail risk to society at large, has been the weakening of our understanding of the price of risk. Now don’t kid yourself: society always stands as the loss-absorber of last resort, under any capital, economic, or financial regime, because there are some losses which are too large for any system to absorb. (Think about a kilometer-wide asteroid hitting New York City or Los Angeles, for example.) After all, financial losses happen to a society. But the drawback of risk assumed by government and taxpayers is that it is not explicitly priced. Leading up to the recent financial crisis, as financial actors’ capital at risk shrank and society assumed ever more tail risk, more and more of the spectrum of possible financial loss fell outside the capital markets’ risk pricing mechanisms. Risk, whether from risky investment projects, financial leverage, or whatnot, looked cheap because an increasing portion of it was not being priced. We all levered up and engaged in riskier activities because we thought those activities had somehow got less risky. Remember the “Great Moderation?” It turns out instead we were just hiding a lot of potential loss out of sight, in the Grandma’s attic of a taxpayer backstop.
I continue to maintain two important things about financial risk. First, that it is incompressible; that is, a certain level of risk is ineluctably tied to the pursuit of a particular level of returns. No matter how you slice it, you cannot reduce the risk associated with a certain return. If it looks like risk is lower than you anticipated based on past history, rest assured it has not declined. You have either transferred it to someone else (e.g., through derivatives, in which case you have transferred some of your return, as well), or you have just lost track of some of it (perhaps by shifting it onto taxpayers). Second, that the risk-return relationship obtains at a society-wide level, as well. It may very well be the case that we unknowingly assumed more risk than we are comfortable with as a society, because we lost track of some of it by shifting it outside the financial markets, unpriced, onto our own backs. We pursued a growth and consumption agenda that was a lot riskier, in retrospect, than we believed at the time. But lowering the risk of loss we are willing to accept as a society will have ironclad implications on the types of returns we enjoy.
Surely there is a happy medium between a low-growth, capital-constrained economy hobbled by unlimited liability to capital providers and the reckless bacchanal we financed with “other” people’s money up to the financial crisis.1 But make no mistake, the decisions we make about how we allocate, limit, and distribute financial risk throughout society—including how much to put financial intermediaries on the hook—will reverberate broadly through the system and ultimately affect our very living standards and prospects:
So part of the conversation we continue not to have in the public domain is what kind of returns—in the broadest sense—we desire for our economy and society, and therefore what level of risk we are willing to tolerate. Sure, we could turn the entire banking industry into a regulated utility, with mandated minimum equity levels, maximum allowed returns on equity, and limits on institutional size and interconnectedness (assuming we can understand, monitor, and control such parameters, which may be a slightly heroic assumption). But what knock-on effects would that have on investors, on businesses in search of risk capital for their growth projects, on consumers, and on the economy at large? Dampen the incentives and ability of financial intermediaries to originate, take on, and distribute investment risk, and it is not clear to me that overall risk-taking (i.e., investment) in the economy will not go down. But if that happens, are we not explicitly or implicitly settling for less growth and fewer wealth creation opportunities in the economy overall? Is that really the outcome we are seeking?
By this, I do not mean to say our current system works well, or that the level of risk inherent in the financial system is appropriate or even efficiently distributed given our overall economic return objectives. But it does mean that we need to be a little more thorough, and a little more honest with ourselves and our opponents in debate, in thinking about the consequences of individual actions or “solutions” we advocate imposing on financial intermediaries. For consequences will flow inexorably in directions we do not—and perhaps even cannot—anticipate, and we will be remiss—and even no less irresponsible than the people who allowed the recent financial crisis to happen in the first place—if we do not make provision to address them.
But cheer up. Things could be a lot worse.
Carolyn Sissoko, A little fear is a good thing (Synthetic Assets, February 4, 2012)
The Blind Men and the Elephant (June 21, 2011)
1 Oops. I guess it was our money all along, after all.
© 2012 The Epicurean Dealmaker. All rights reserved.