Because I have yet to throw my stuff back in my bags to be on a morning plane back to NY, I must be brief.
I have not had the chance to think about Avinash Persaud regulatory proposals at VoxEU deeply, but at a first glance, they sound appealing. However, the devil will lie in the details, particularly since all his programs would be new. For instance, he suggests a new axis for distinguishing between lightly and more heavily regulated institutions. While I like the idea, any sharp line will encourage regulatory arbitrage; a gradient might work better. And these proposals are so far from the traditional approach that their novelty alone will provoke resistance.
From VoxEU:
Financial regulation never works the way it should. Here one of the world’s most experienced analysts of the global financial system presents some remarkably clear thinking on why we should not just do more of the same. An alternative model for policy action is proposed.
I have had the misfortune or fortune of being up close and personal with seven major financial crises in my banking career, from the US Savings and Loans crisis of the late 1980s to today’s credit crunch. In each crisis I have observed a “cycle” in the response to the crisis. In the middle of a crisis, when circumstances look dire and chunks of the financial system are falling off, proposals get radical. I recall in December 1992, with the UK and Italy having already been ejected from the European Exchange Rate Mechanism and Spain and Portugal looking vulnerable, some European policy makers flirted with capital controls. But a few months after each crisis is over, these radical plans are tidied away and we are left with three things. And they are always the same three things: better disclosure, prudential controls and risk management.
These measures are the regulatory version of apple pie and ice cream. Who would say no? The thing is – we have been investing heavily in these areas for the past twenty years and do not have much to show for it in terms of financial stability. Over the past eleven years we have had the Asian Financial Crisis, LTCM, the “dotcom bezzle” and now the credit crunch. While more disclosure, controls, and risk management are generally good things and necessary fraud reducing measures, there are few crises I have known from the inside that would not have happened if only there was more disclosure. People knew that sub-prime was a poor risk – it is called sub-prime, after all.
Regulatory shortcomings
The problem is more fundamental, and, unless we address these fundamental issues, we will be condemned to repeat the cycle of boom and bust. Lying close to the heart of the problem in all of these recent crises, from today’s credit crunch to the Savings and Loans debacle and beyond, is the inappropriateness of financial regulation.
My own view of banking regulation would be considered quaint next to today’s practice. I consider the primary objective of intervening in the banking market to be mitigating the substantial systemic consequences of market failure in banking. It is therefore puzzling to me that market prices are now placed at the heart of modern financial regulation, whether in the form of mark-to-market accounting or the market price of risk in risk models. It is not clear to me how we can rely on market prices to protect us from a failure of market prices. I have discussed this before many times so I will focus on the secondary objective, which is to avoid the discouragement of good banking.
A good bank is one that lends to a borrower that other banks would not lend to because of their superior knowledge of the borrower or one that would not lend to a borrower to which everyone lends because of their superior knowledge of the borrower. Modern regulators believe this is too quaint, and, to be fair, many banks were not any good at it. But instead of removing banking licenses from these banks, regulators decided to do away with relationship banking altogether and promoted a switch away from bank finance to market finance where loans are securitised, given public ratings, sold to many investors including other banks, and assessed using approved risk tools that are sensitive to publicly available prices. Now, bankers lend to borrowers that everyone else is lending to, the outcome of a process where the public price of risk is compared with its historic average and a control is applied based on public ratings.
Market finance
This switch to market finance improved “search liquidity” in quiet times. Credit risk that was previously bundled with market and liquidity risk was separated, priced and traded. This has improved the transparency and tradability, but it comes at the expense of systemic liquidity in noisy times.
Almost every economic model will tell you that if all the players have the same tastes (reduce capital adequacy requirements) and have the same information (public ratings, approved risk-models using market prices) that the system will sooner or later send the herd off the cliff edge (Persaud 2000). And no degree of greater sophistication in the modelling of the price of risk will get around this fact. In this world, where falling prices generate more sell-orders from price-sensitive risk models, markets will not be self-stabilising but destabilising and the only way to short-circuit the systemic collapse is for a non-market actor, like some agent of the tax payer, to come in and buy up assets to put a floor under their prices. (I wrote about this liquidity trade-off with some colleagues; Laganá et al. 2006)
Now this is a legitimate model: the marketisation of finance and the resulting improvement in search liquidity in quiet times, coupled with direct state intervention in the crisis. It is the model we have today. But I venture that it is a highly dangerous model. It is expropriation of gains by bankers and socialization of costs by taxpayers. Paying for a decade of bank bonuses can be very expensive for the taxpayer and the opportunities for moral hazard are enormous.
An alternative approach
The alternative model rests on three pillars. The first recognises that the biggest source of market and systemic failure is the economic cycle and so regulation cannot be blind and deaf to the cycle – it must put it close to the centre. Charles Goodhart and I have proposed contra-cyclical charges – capital charges that rise as the market price of risk falls as measured by financial market prices – and a good starting point for implementation of such charges is the Spanish system of dynamic provisioning (Goodhart and Persaud 2008).
The second pillar focuses regulation on systemically important distinctions, such as maturity mismatches and leverage, and not on out-dated distinctions between banks and non-banks. Institutions without leverage or mismatch should be lightly regulated – if at all – and in particular would not be required to adhere to short term rules such as mark-to-market accounting or market-price risk sensitivity that contribute to market dislocation. Bankers will argue against this, saying that it creates an unlevel playing field, but financial markets are based on diversity, not homogeneity. Incentivising long-term investors to behave long-term will mean that there will be more buyers when banks are forced to sell.
The third pillar is requiring banks to pay an insurance premium to tax payers against the risk that the tax payer will be required to bail them out. If such a market could be created, it would not only incentivise good banking and push the focus of regulation away from process to outcomes, but it would provide an incentive for banks to be less systemic. Today, banks have an incentive to be more systemic as a bail out is then guaranteed. The right response to Citibank’s routine failure to anticipate its credit risks is not for it to keep on getting bigger so that it can remain too big to fail, but for it to whither away under rising insurance premiums paid to tax payers.
7 comments
Comments are closed.
How about compartmentalization at the state level (keeping banks more local)?
Persaud’s proposals merit more, and more general, discussion they they will likely receive in the near term, not least because they are concrete and ones which could be implemented at the discretion of public authorities. I like all of them in principal, but the specifics are not clear and do indeed matter. I’ll raise additional proposals in a separate post.
Pillar One: Contra cyclical dynamic provisioning requirements for financial institutions would be most welcome. The index for the ratchet is what counts, though. Economic cycles do vary in length and focus of effect, and more significantly are not easily predictable with regard to their peaks. Where the cycle turning points are designated, and especially by whom, have major cost impacts for financials, and so such decisions and their makers should be well insulated from influence. This suggests that a stiff ratchet should be used, i.e. that the number of years (or better quarters) since the last cycle turn should be a stiff multiplier for driving reserve provisioning levels.
Pillar Two: In principle, it is no bad idea to grant less support and hence increasing speculative latitude to financial players with little leverage and minimal term mismatch exposure—all three of them, as there just won’t be many players of any size that fit that profile. However, the _lack of systemic risk_ carried by such players, i.e. low leverage &etc., needs to be verified by public authorities. Just taking their word that “We’re good for it” doesn’t cut it. And their needs to be clear criminal penalties for lying to public authorities. There also seems to be an underlying assumption here, perhaps more in Persaud’s phrasing than in the concept, that such large, wholly owned speculators shouldn’t be held to short term rules because they are buying into long-term positions. The opposite seems often to be the case in recent times: their positions are often hot and fluid, or meant to be. Hot, fluid money is by no means necessarily or at all a healthy things for smart, safe economic development, often the reverse. This is the real cost of letting big private bucks speculate as they please, they burn other people’s interests in the process. But focusing on outcomes rather than regulating processes here is perhaps a cogent approach.
Pillar Three: Charging financial institutions a Systemic Insurance Fee is a fine idea. Scaling said fee up increasingly relative to their size is even better. However, let’s not leave out the concept of capping financial institution size by assets, by exposure, and likely by both. Currently, we don’t really do this, although there are weak limitations on market concentration. But really, behemoth banks or quasi-banks give few advantages to society. It may be as you say, Yves, that banker-brokers must be of a certain size to capture enough of the market to maintain a viable position. If that leaves the public exposed due to the size of their risks the situation all but cries out for public guarantees and oversight, regardless of how, exactly, such are conceived. The banking industry, like the airlines, has demonstrated that yes, it WILL kill it self if left to its own devices; the principle difference is that banking systems have _repeatedly demonstrated this capacity historically whereas the airline industry has only done so twice.
We needn’t worry about any of these proposals being passed soon, though. It was five years into the Great Depression before serious financial reforms were implemented although public intervention to prop up the markets was pursued from the outset. We can’t even agree at the present that we are in a crisis; “I mean, it’s over already, isn’t it?” The first public policies we see will be either window dressing, blanket protection proposals for existing stakeholders, or dithering. We’ll have time to shape up reasonable proposals before cycle trough political pressures build to the point of pushing through real reforms.
How did the financial industry come to the pass we now face? This is the first question to ask in considering what structural or regulatory changes are desirable. The fundamental issue, to me, is the unwillingness of firms lending money to set aside appropriate reserves against losses, at any level. We have 300 years of modern banking history which has without exception indicated that unreserved lending is to a financial institution what the absence of an immune system is for an organism; a scratch can kill you (default cascade or credit cut off), while a real virus not only kills you but infects your neighbors. So we see again. This behavior, an unwillingness to reserve against losses, suggests its own trajectory of solutions but let’s do a brief review for context.
Loan retailers, including mortgage brokers, set aside very little for losses because they weren’t going to hold the debt; instead, they pushed it up the chain, typically for securitization. Banks skirted their reserve requirements by opening conduits with pitiful liquid reserves to park debt of various kinds while shopping it or bundling it to be shopped. Similarly, banks underwrote huge volumes of inherently risky and unstable LBO debt against which they compiled no adequate reserves because, again, they expected to sell the debt at a profit not retain it.
CDOs are the freak show exhibit for tortured ill-thinking about how to reserve against losses. The principle benefit, initially, from securitization was overcollateralization against losses. Yes, really. This had at least three legs, of unequal size. In many cases, default swaps were bundled into the CDO as a shock absorber to take first losses. The CDO was sold at a discount to the face of the underlying debt, so that a further cushion against loss was bundled in. Both of these provisions were unequally distributed to tranche buyers, but in principle offered significant reserves against loss risk. Finally, some CDOs had limited recourse provisions against the originators of the original debt in case of fraud, high failure rate or the like. These mitigation options were small and hardly universal, but again they in principal reserved against risk.
All of these ‘reserves’ have failed massively in the present circumstance, and for much the same reason: they weren’t real reserves—cash or near equivalents tied to the debt—but promises of payment. Issuers of default swaps as we see never expected to pay off more than a tiny fraction of their swaps, and to the extent that they themselves had any ‘reserves’ these proved to be not cash but debt which in a pinch they have been unable to sell to raise money. The swaps on any one CDO may pay out, but on the instruments as a whole _cannot_ pay out. Then the underlying debt bundled in CDOs has tended to be overconcentrated in single asset classes, and thus totally, even ridiculously, exposed to price declines in the same asset class. The ‘excess collateral’ has been wiped out and far more by overall price declines. And many loan originators have simply gone out of business, or are accidents awaiting liquidation, eliminating pittance mitigation from retail underwriters. There were no REAL reserves in these CDOs, only promises to pay. This is the biggest fallacy of passing risk around the financial system, that promises without substance will be honored, or even can be.
Banks lent a great deal of money against which they retained no reserves. This in fact was a principle accelerator of the bubble in asset prices, because these hot, fluid, expanding vaporbucks competed for the same assets and so inflated their prices. This had the appearance of inflating asset _values_ but this was not really the case. Hopes that actual gains in asset values would cover any potential (and putatively unlikely) losses proved utterly speculative in all the worst sense of the word. Thus, at the same time that banks contributed to a balloon of asset prices they underreserved against the risk of trafficking in and owning those same assets, in effect multiplying their exposure to loss.
The public authorities also failed to reserve against risk in this, even leaving aside their regulatory dementia in allowing banks to vastly expand their exposure without increasing their reserves. The authorities did this by their implied, and now explicit, guarantee to let no major institution fail. With that hope and belief, why would big institutions lending money hold _any_ reserves, let alone large ones? And while the Fed had 800 gigabucks to play around with here, and many regulatory fudges, that sum isn’t nearly large enough to backstop the entire financial economy of the US. So the Fed didn’t really have the money to put where their mouth has been, either.
The issue isn’t simply that the financial system, in whole and in part, took excessive risks. Far more, it is that they system and all it’s players convinced themselves they didn’t need to set aside money commensurate to the amounts they were moving around—because the ‘vaporbucks’ would always stay in motion until they ended up somewhere else. We need to return to the concept or requiring solid and sizable reserves against losses for parties that lend large amounts of money.
And those reserves at the Federal ‘Reserve’ System? Well, part of them need to come from increased fees levied on regulated players. However, we also need the option of nationalizing failed or failing institutions, wiping their equity holders and shaving their bondholders to the extent necessary. We need that option in part to keep the public authorities from being greenmailed by financial institutions or cartels of same ‘too large to fail.’ The public needs to be able to give failing marks to those that so merit and run them out of the game. And money set aside for the purpose in the hundred billion dollar range needs to be there so that the threat has substance.
A good bank is one that lends to a borrower that other banks would not lend to because of their superior knowledge of the borrower or one that would not lend to a borrower to which everyone lends because of their superior knowledge of the borrower.
One question that springs to mind upon reading this is: of what, exactly, does this suppoosed “superior knowledge” consist? It would at least seem that anything that can be objectively measured is something that cannot be the private province of a particular bank. At least not in principle: of course a borrower may choose to share details of his financial situation only with his lender, but such is not some sort of special knowledge on the part of the lender, since the same information could be provided to another lender, if the borrower chose to work with another lender.
But if this is so, then the only “superior knowledge of the borrower” that a lender could have would be either access to information that was inappropriately disclosed (or not disclosed), or subjective personal opinions, both of which seem to be ripe for abuse.
“When the capital development of a country becomes a by-product of the activities of a casino, the job is likely to be ill-done.”
–JM Keynes
Current Fed actions are not resolving the crisis, they are postponing it. Keep injecting reserves, keep swapping Treasuries for toxic waste, and eventually all that happens is the I-banks swap their MBS to the Fed for Treasuries and with them safely locked away, they channel their newfound liquidity into large cap equities and commodities. That is EXACTLY what is happening. Now that the Fed is willing to take the garbage off their hands (effectively for as long as they are willing to rollover the TSLF positions) Wall Street can go about blowing new bubbles in equities and commodities. None of that changes the fact that the toxic assets still exist. What happens when the Fed decides to stop rolling over the TSLF loans? What happens when (due to inflationary pressures) the Fed has to raise the Fed Funds rate? I still only see a few options – Great Depression style crash, decade long (or longer) zombie economy ala Japan, or hyperinflation to ameliorate the debts. There is no soft landing option, near as I can see.
Instead of using superior knowledge, he writes, “Now, bankers lend to borrowers that everyone else is lending to”
Persaud offers some intriguing suggestions. Still, any conversation about his or anyone else’s proposals will be more productive if language is used in a meaningful instead of nonsensical fashion.
Contrary to his description, banks were not ‘lending’ during the heyday of the securitization and ‘innovation’ scam. Banks were ‘handing out money to all comers’.
Any regulatory scheme will fail if it permits banks to ‘hand out’ money instead of ‘lending’ money.
That’s pretty basic. But, then, the three card monte wizards of ‘innovation’ prefer banks that hand out money over banks that lend money.
The principle should be: too big to fail. If a company is so big that it’s failure will disrupt the larger economy that it will have to be bailed out by the government then the company must submit to a higher level of regulation and oversight.