With the financial system on the exam table, it has been more than a bit troubling, that certain questions are neglected in serious academic/policy debates.
The discussion of possible remedies focuses on regulatory solutions, everything from requiring mortgage brokers to be licensed to increasing financial institution capital requirements and having much greater harmonisation, as the Brtis like to put it, of banking and brokerage firm oversight.
While these measures individually and collectively could be salutary, no one seems to be willing to consider the fundamental question: did the push to facilitate the free flow of capital, both domestically and across borders, play a role in this crisis? For the last 15+ years, the push in policy has been towards increased efficiency, which means lower transaction costs, less supervision, little interest in considering whether so-called innovations benefit anyone beyond their purveyors (Martin Mayer observed that, “A lot of what is called innovative is simply a way to find new technology to do what has been forbidden with the old technology.”).
It’s important to examine this question, because many in this society have come to believe that regulation is bad and ever to be avoided. Yet markets like the equities markets, where participants trade an ambiguous promise anonymously, depend on regulation. Thus, the question should be, “What level of regulation is optimal?” rather than “How much regulation can we eliminate?” The problem with the latter approach is that it can take years for problems to develop, and when they do show up, since the tools to stop them have been thrown away, a full blow crisis has to develop for corrective measures to be implemented.
One perspective on this issue that has generally been given short shrift is thinking about the financial system as a system and looking for lessons and analogies outside the realm of money.
Reader Richard Kline has been pondering this issue in a series of posts (see here, here, here and here) from a complex systems perspective rather than the traditional finance/markets vantage point. The discussion below summarizes his argument; a fuller treatment can be found here.
In this post, Kline addresses possible solutions. As you will see, viewing the problem from a different axis leads to some fresh observations (some run the risk of eliciting outrage) that will hopefully stimulate thinking and comment.
From Richard Kline:
In speaking of ‘remedies’ for financial system instability or outright speculative excess, the images which come to mind are regulators in tired suits and thick spectacles, and statutory limits and penalties hidebound in Moroccan. To be sure, the financial systems needs such hard limits. Fundamental contributory factors to the present financial crisis include insufficient and imprudent capital reserves and excessive leverage. Both factors are directly contributory to every bubble you will ever research, are known as such, and act in wholly linear fashions. Harder and higher statutory and administrative limits including mandatory public oversight of capital and exposure are required because participants simply won’t act with restraint without regulation; we see it now as we have seen it time and again.
Beyond this, in systems in general, and the financial system as a specific example, high connectivity is associated with undampened propagation, massive correlation across nodes, and tight couplings between subsystems. All these outcomes are known to diminish systemic stability through cascade effects. Cascades can certainly be limited by outright compartmentalization such that portions of the system communicate through overtly controlled ‘gates.’ This is highly inefficient however, and moreover difficult to achieve in the case of the financial system which is highly fragmented into many nodes. Soft compartmentalization may be more effective, especially through constraints which scale.
In systems in general, self-modulation is evidently associated with throughput overshoots of nodal capacity, and potentially of systemic capacity. Hard limits on the volume of throughput are again inefficient, and moreover will be hard to enforce in the global financial system. Concentrations of throughput—capital, debt, and risk—need to be closely monitored, but systemic connectivity must be considered in tandem due to its capacity to shift stress from system locations of obvious concentration. In general, interventions which serve as ‘capacitors and resistors’ to throughput may have the largest potential value.
Financial markets do have a history of developing effective ‘prudent practices’ which require no centralized intervention. Good examples are margin calls, asset diversification, and subsidiary liability firewalls. These examples: a) all apply at the point of exposure concentration; b) are locally triggered rather than centrally activated; c) require offsetting action without mandating investment purchase or sale; d) generate a verifiable (and certifiable) paper trail, or should; e) residually promote system stability through braking exposure concentration, which to a degree forces differentiation and hence discorrelation. By no means all structural remedies can fit this profile, but together they suggest an optimum design where applicable. The following are several broad concepts of how to implement controls in a many node system, from which specific policies could be drawn.
Dynamic constraints: The optimal form of a ‘control’ in the financial system would be trend active, threshold passive, and nodally focused. That is, when some variable of local concentration reaches a pre-set level, say the volume of a certain asset held, or the shift in value of a certain obligation, a mandatory adjustment is initiated. The thermocouple in your thermostat does this: it automatically trips off, and should you disagree you must overtly act to override the setting. A margin call or a stop-loss sell order both function in this way. In principle, many financial practices are made to look like this, such as mark to market accounting or regulatory reserve capital; in practice, those with the open position have considerable practical discretion regarding when to act, i.e. they are not tightly coupled to re-set. Implementation must be formally automatic, with any steps to delay or cancel being overt and subject to approval.
Now, hard sudden braking at a threshold can be disruptive, both locally and systemicaly; this is why those lending on a margin may be understandably reluctant to call, or slow to act. The solution to this is two-fold: a) more frequent but initially smaller restraints which, however, b) scale progressively larger. Thus, for example, loss reserves should not just increase proportionately at large intervals but rather increase exponentially but gradually over shorter intervals. Leverage may be high for small volumes, but scale down exponentially against large volumes. The onset of a dynamic constrain best acts more as a warning than a brake, but it should promptly build to more a brake than a warning.
Modulate flows: Modulation of throughput rather than outright barriers offers a high return of effect for a low investment of cause. However, interventions must counter-trend. For example, the more money lent or borrowed for involvement in a single market, the higher the interest rate should be for successive increments. This acts against maximum profit, and so seems counterintuitive; however, this also acts against concentration of loss in one node, and correlation of risk in one type. This is ‘inefficient’ but more stable: notice the direct relationship between those conditions. As another example, the larger a position, the less one may hedge (or insure) it and the more one must reserve it. The more of a specific product one moves, the greater the duration between further offers of set volumes of the same. Modulation can be scaled back in a like fashion. Actions to modulate flows can apply node by node, but these also can be used at the macro level by central banking institutions or clearinghouses of various kinds. The point of modulation is less to bar actions outright since ‘the perfect level’ or risk/reward is often ambiguous—or shifts. Rather, the goal is to circumvent faulty guessing about perfection and instead raise costs and lower velocity steadily to dissuade further involvement (and the reverse on the downswing).
Scale flows: Research with many biological systems indicates that they maintain overall stability by power-scaling structures of common form but varying size, often by quarter-power exponents. The applications of these findings need wider consideration for financial system nodes and flows, but two observations serve here: a) restraints on flows must scale exponentially and progressively rather than as now proportionately and abruptly, as mentioned above; b) different levels of constraint should apply at different levels of the system. For example, after adjusting for risk the same interest rates in principle apply to ibanks and used car purchasers. Power scaling implies that participants of smaller scales should be able to borrow more cheaply than those of large scales. There is secondary evidence which implies this in that small borrowing is known to better stimulate overall demand while large borrowing manifestly concentrates risk. Again, small investors can concentrate everything in a single asset without risking system stability, while this is less true of very large investors.
The takeaway on scaling flows and stability is that the larger the concentration of capital the more it should face higher costs and constraints whereas at present the reverse is true. This is a major reason why open capital systems are inherently unstable: they scale the wrong way.
Dis-correlation: Large concentrations of capital concentrate risk. They also tend to correlate investment. They have maximal connectivity on the whole, and so propagate those risk factors systemically more than do smaller nodes. They are minimally adaptive to systemic change. If they cannot necessarily achieve outright monopolies—a Bad Thing unless closely regulated, and often even then—they have the potential to squeeze prices. Moreover, few large participants are more likely to distort flows or markets, though self-correlation alone if not outright collusion. When truly large, they create node-dependency within the system. It is difficult to find a single countervailing positive for large capital concentrations, let alone an offsetting advantage set.
Market de-concentration needs to be strictly enforced, both by market participants and by overt regulation, if markets and the financial system as a whole are to retain ongoing stability; this cannot be over-stressed. Firm sizes, market share limits, asset concentration, and the like all require monitoring and strict caps. Current large players will hate this, but we all need it. Where large concentrations of capital are needed, syndication can serve quite well. Syndicates are likely to be slower to move, and are special purpose rather than ongoing: these are desirable outcomes for system stability, features not bugs. In a nutshell, empires are bad, city-states good (or at least less bad).
Mid-term counter-cyclical restraints: Modern macroeconomic flows are profoundly pro-cyclical, as several commentators have argued. Investor-speculators are too loose and liquid on the upswing and too tight and sticky on the downswing. Prolonged expansions tend very much to concentrate investment and cumulate risk; both developments have undesirable outcomes. In principle, central banks act counter-cyclically, but in practice they are successfully pressured to be slow and timid on the upswing, and fast and reckless on the downswing; i.e. they partially correlate with pro-cyclical movements even when acting against them. Moreover, central banks have discretion to act, can get it wrong, and are inordinately swayed by exogenous impacts; in consequence of these distortions, they typically act too late.
Beyond this, markets which optimize for short-term outcomes may be inherently exposed to mid-term divergences, and so optimize themselves away from long-term stability. Bankers, central and otherwise, famously do not optimize for the long-term. What is to be done, went the famous question?
It is too little understood that systems which oscillate are stable systems; put another way, the financial system can ‘expand’ continuously or it can be stable but not both. One remedy, imperfect but impactive, would be for macroeconomic policy to adopt the same schema of dynamic constraint suggested above: trend-active, threshold passive, scale to exposure, overt override. Small but certain and regular increases in top tier interest rates, mandated reserve increases, and required deconcentration begin with every upswing. The same reverse with every downswing. Central bankers and other regulators if they certify the need (a brake on too ready intervention) can act to stay or accelerate such trajectories—but not to reverse them. Some flexibility in timing would likely be desirable, but not much. Exogenous events will produce salient changes in trend, and one might leave a ‘declared emergency’ option to respond to abrupt trend reversals—but only by moderate increments. Markets and investors would have a clear profile of forward macroeconomic intentions. Turning points are not clear, and there is no perfect formula for this. Perhaps such an ‘invisible governor’ could skip an increment if the inputs are unclear, but must re-set one way or the other on the next re-set term. The advantage of this approach is that it can be driven by logistic trend deflection points in flows, concentrations, correlations, and duration of trend against historical norms which are typically evident before severe asymmetries have time to develop.
Such a concept again seems counterintuitive; central bankers are ‘supposed to act.’ The evidence is that they tend to miss the mark, and that in attempting to support the impossible—permanent expansion—they act irrationally against system stability time and again. Dynamic counter-cyclical action can act gradually where at present central bankers and political actors can’t bring themselves to act. There is the lesser implication that official rates should move away from quarter or half point intervals to decimal ones which have finer resolution, but which through greater variability may serve as a moderate constraint on financial market anticipation.
Compel savings: All debt and no savings means that there’s Hell to pay; call it Faustian capitalism. So how to induce savings? Requiring everyone to pay into a 401k builds better brokers’ bonuses but delivers rather less than advertised over the long-term. Moreover, this actively hurts banks, and inflates ‘liquidity’ in a way that is likely systemically destabilizing. Besides, large net worth participants don’t save anyway: they invest, i.e. more money in motion where less may be ‘more better.’ Further, the investor class cannot be induced to put their money into social needs voluntarily, especially those with long time frames, such as infrastructure, education, and low income housing.
It is time to consider the return of the forced loan. Above a set level of taxes, further premia should be assessed against income and gains where money is borrowed at nominal interest rates and repaid at or over a set term. Payees having the option to roll over their funds, perhaps with a tax incentive. This should start with high net worth, but eventually directed loans should be assessed against all earners in some amounts; whether to apply this to businesses is debatable. Developed in antiquity, the forced loan was a favorite of pre-banking societies; it’s principal use was to finance warfare, an admittedly bad precedent which should be specifically barred by statute. Furthermore, government falls in love with income sources, and revenue flows grow constituencies, so care needs to be given to the destination of such capital. Most certainly directed loans should be partitioned from general fund expenses.
Bonded sovereign debt has a critical function in the capital markets; no new scheme should replace a significant role for public debt. Directed loans cannot ‘replace’ taxes anymore than debt can replace revenue. They can serve to finance long-term, low risk social needs rather than borrowing on the capital markets at higher cost. More importantly, directed lending may serve to shift some societal capital away from speculation at the high end and consumption in the great fat middle into capitalizing the social and financial systems indirectly. This concept needs more development than it will receive here, but it can promote system stability while having some individual benefit.
Relational models: Discussions of the financial system are relentlessly cause and effect in reference. However, cause and effect does not operate the same way in systemic relationships . . . It is more accurate to speak of modulation, catalysis and emergence. Inappropriate conceptualization reliably yields poorly fitting analysis and false conclusions. Much of current macro-economic reasoning is defective because it is overly concerned with lower dimensional supply and demand factors rather than inherently higher dimensional systemic flow and concentration factors; more accurately such reasoning is only locally effective, not globally (in both the figurative and literal senses). This is why we presently see neoliberal macroeconomists inside the Fed and out jamming away at the ‘demand stimulus’ button on their adding machines with puzzled looks on their faces while nothing happens where they are but the system shorts out elsewhere (presently behind the NYMEX panel). If and as economists use models, we need to see more of long-running relational models (that is a technical term) and less of short run linear models. This should be a focus of government level statistical research.
And while we’re at it, good models need good statistics. Going forward, it would be better to concentrate the compilation of critical statistics away from regulatory agencies and Executive control specifically and toward neutral agencies without larger policy briefs. The GAO and even more the Comptroller of the Currency come to mind. No remedy will do much good if it’s cut with baby laxative while someone’s thumb is on the scale.
James H. Brown, ed. 2000. Scaling Laws in Biology.
John H. Bodley. 2003. The Power of Scale: A Global History Approach.
[The papers in Brown’s collection have important implications for systemic organiztion. Bodley’s text is an anatomy of how scale factors adversely structure political economy, an invaluable work.]