It may seem odd that a relatively upbeat assessment of the prospects for securitization in Latin America (by Michela Scatigna and Camilo Tovar) has been released this month in the BIS Quarterly Report, while securitizing loans and creating derivative products are being blamed as main culprits for the current financial mess in advanced economies. But I am among those who believe that the financial innovations of the last few years have come to stay – however fixed after the crisis. Thus, Latin American economies may have a chance to benefit from lessons painfully acquired elsewhere as they are still giving the initial steps toward securitization of their own assets.
The Bright Side of the Moon
Asset securitization corresponds to the transformation of illiquid assets (non-marketable assets or future cash flows) into tradable securities, named asset-backed securities (ABS), either by selling the underlying assets or by purchasing protection against slices of their risks. The originator of ABS is able to anticipate cash inflows or manage better the risk profile of its balance sheet.
Individually illiquid assets are pooled into portfolios, what diversifies away the dependence on an individual asset’s performance. These pools are usually transferred to Special Purpose Vehicles (SPVs) in order to single out the assets/risks from the originator’s balance sheet, for both legal and managerial reasons. “Synthetic securitization” is the expression used when it is the credit risk of the pool of securitized assets that is transferred by means of credit derivatives, rather than the ownership of assets.
In the case of grouping assets for sale, some kind of “credit enhancement” (reduction of default risk) is usually offered, via over-collateralization (security issuance lower than asset values), purchase of credit default swaps or other partial risk guarantees, negotiation of lines of stand-by credit, and others. The asset pools can be additionally divided into several classes of other securities (“tranches”), to which one may attribute different risk levels. As in the “sequential pay” structure, in which the tranches are entitled to retirement in sequential order, from “senior” and “mezzanine” to “junior” (“equity”) ones, with correspondingly differentiated levels of credit risk.
From the originator’s standpoint, the potential attractiveness of ABS as a tool of risk management is straightforward. It can use the improved risk profile either as an objective in itself and/or as a larger risk space for augmenting its degree of portfolio leverage. In the case of banks, asset securitization widens the scope for regulatory arbitrage and may lower costs of compliance with capital requirements.
From the asset purchaser’s – or risk-protection seller’s – standpoint, ABS may also become a way to diversify risk exposures. It is worth noticing how ABS may be a way to mitigate difficulties associated with information asymmetries that typically plague some types of assets: e.g., pension funds become capable of investing in long-term residential assets by acquiring senior tranches of mortgage-backed securities (MBS), whereas in the absence of these securities housing finance becomes limited to loan origination by those financial agents close enough to clients (banks, mortgage companies).
From the systemic standpoint, asset securitization can be seen as a move toward “completeness of markets”, creating vehicles for risk sharing and transacting otherwise unattainable. Therefore, in the absence of some other aspects that only now have become clear (approached later), the net effect of securitization tends to be more and cheaper credit throughout the whole financial system. There are also potential positive externalities, i.e. additional benefits that come beyond those gains accrued individually by issuers and purchasers. This is the case when better risk-managed balance sheets lead to higher systemic resistance to shocks.
Securitization in Latin America: starting to shine
All those are potential gains from asset securitization that remain far short from being explored in Latin American economies. Costs of funding and intermediating risky assets might be lower in many countries of the region if the scope and depth of financial markets had already benefited from a higher extent of asset securitization. More widespread transformation of illiquid assets into pools of liquid ones might have helped increase the supply of finance in many areas where the gap between what borrowers are able to offer and what financiers are willing to accept is typically high (housing finance, long-term corporate loans and other receivables). But according to the BIS paper, some signs of initial progress in that direction can already be noticed.
The expansion of securitized transactions in the region has been substantial in the last five years, particularly at the domestic side: over 80% of the total US$20 billion issued in 2006 were in local markets. It is still a small size when compared even to other emerging regions, but significantly higher than the regional US$ 6 billion of 2002 – these and other figures below can be found in Fitch (2007).
Prior to the crises of the second half of the 1990s, securitization in Latin America consisted mainly of simple cross-border transactions (foreign currency-denominated securitization of receivables of exporters and of a few large, financially sound and highly creditworthy local originators). Since 2003 many second-tier originators, unable to access international sources, have increasingly resorted to securitization in domestic markets as an alternative less costly than traditional bank finance. Commercial and residential mortgages, auto and consumer loans and trade receivables have all increased aided by newly created possibilities of securitization, made possible by macroeconomic stability and legal initiatives in some countries. Sub-investment grade structured issues have already taken place.
On the other hand, these recent developments have been particularly geographically concentrated. Chile, Colombia and Peru, for instance, have all progressed in terms of asset securitization in relative terms to the size of their financial sectors, but in fact Brazil (US$ 5.3 billion) and Mexico (US$ 6.5 billion) accounted for 72% of the regional local market issuance in 2006, with Argentina (US$ 2.4 billion) coming third. According to Fitch, prior to its crisis, Argentina had presented an early spurt of structured transactions.
Moreover, outside from Brazil and Mexico, just one kind of asset predominates: in 2006, credit cards represented 45% of Chilean transactions, residential MBSs were 60% in Colombia, and future flows 46% in Peru. In Brazil, for the first time a Collateralized Debt Obligation (CDO) was issued last year, whereas payroll-deductible personal loans, auto loans, credit cards, utility bills and commercial flows have led the process of asset securitization. The BIS paper highlights that most of the deals are now placed in the public capital markets, rather than using private placements as before. Residential and commercial MBSs corresponded to only 9% of all structured transactions and still have a lot of potential to develop. Mexico has advanced farther in residential MBSs, as well as in placing other ABSs denominated in local currency.
Besides the obvious scope for higher housing finance, a need so neatly unattended in the region, ABSs may also contribute to corporate bond issuances, as long as the corresponding markets cease to be restricted to first-tier and highly rated firms and minimum scales for pooling assets can be reached. Improvements in legal frameworks and incentives/requirements toward higher information disclosure by private companies will also be necessary.
Overall, the region would seem to be poised to exhibit progress in asset securitization. Nonetheless, judging from what is happening in the US financial sector, would that be good news?
The Dark Side of the Moon
“Credit turmoil shows not all innovation has been beneficial” was the title of an article by Gillian Tett in the Financial Times of September 11, expressing the mood of disenchantment with structured finance that has followed the current financial crisis. After all, as the crisis unfolds, widespread asset securitization can apparently be blamed as having led to higher financial instability in the US and abroad, rather than to the potential positive externalities in terms of higher resistance to shocks.
The burst of the US housing bubble ended up dragging down first the layer of outstanding sub-prime loans, the large set of which could not have been made possible were it not for the use of ABSs that allowed loan originators to transfer sub-prime risks and leverage their own positions. Then a crisis of confidence on the soundness of the whole fabric of ABSs and on the immunity of banks followed suit very rapidly, with an impact proportional to the huge dimensions that structured finance had acquired in the system.
In fact, there were potential failures in the move from the originate-and-hold to the originate-and-transfer models that had been underestimated:
1. Complexity. As The Economist (September 22, p. 86) said it, “The boom in derivatives was one of those moments when financial engineering raced ahead of back offices and risk-management departments, leaving them struggling to value or account for their holdings. Pierre Pourquery, of Boston Consulting Group, says it is not uncommon for investors to break their exotic purchases into smaller pieces in order to feed them into their risk-management systems. This brings new risks, particularly that the parts will behave differently from the whole under stress. (…)”. Even in the case of simply bundling loans into securities, debt instruments that were accurately priced before were transformed into new ones the pricing of which was too much dependent on arbitrary assumptions. As a moment of lack of confidence and propensity to liquidate assets was sparked, the perception of unclear pricing criteria had an effect similar to throwing fuel on fire. 2. Lack of transparency. The complexity of transactions led to hidden connections that were not accounted for. Risks were placed with investors that were neither knowledgeable nor fully aware of them, such as e.g. those typically conservative money market fund investors who acknowledged to have underestimated the implications of holding exposure to sub-prime loans. Furthermore, techniques of “credit enhancement” ended up turning securitization not as arms-length as announced, with off-balance-sheet connections not taken into account. The downside of risk spreading was a lack of information about the location and intensity of potential credit losses, as the BIS repeatedly hinted at for a couple of years.
3. Dilution of responsibilities. Complexity and lack of transparency were combined with distorted incentives for risk underestimation. The most conspicuous case is the easiness with which rating agencies attributed high grades to mezzanine and junior CDO tranches, especially when capturing advisory fees from issuers, what ended up obliging them to make sudden corrections at too late stages. Furthermore, in general terms, the divesture of mezzanine and junior (equity) tranches by issuers went too far, weakening the incentives for appropriate screening of loans at the origination point.
4. Euphoria. This has been a feature common to all initial diffusion stages of major innovations in the history of market economies, periods in which the understanding of innovations themselves is very limited and “greed overcomes caution” by far. The three previous ingredients of failure were seasoned with a euphoria that led to over-lending, overly low cost finance and over-securitization. Tighter monetary policies in the last few years would have hardly been capable of fully containing such euphoria.
5. Unexpectedly high financial contagion. The extent of over-securitization and generalized over-leverage, in a context of complexity and obscurity of exposures, crossed the tipping point after which the stabilizing effects of risk dispersion and individual portfolio diversification were surmounted by the overall financial instability generated by simultaneous massive asset liquidations.
While the net balance of both boom and bust stages of the recent cycle of securitization is still to be determined, as the financial crisis is still unfolding, one bet seems inevitable to us: ABSs, MBSs, CDOs and other devices of risk transfer will be mended, but not disappear. As examples of possible aftermaths, I would point out:
(i) As it has happened with all innovation cycles in the past, the depressive mistrust that succeeds euphoric bubbles will also phase out after a certain moment, leading to a more stable path of expansion and more balanced proportions of structured finance in the system.
(ii) More standardized and simple-to-understand products, often issued in public placements, will increasingly substitute current complex and obscure over-the-counter transactions.
(iii) Requirements regarding transparency of exposures will become more stringent.
(iv) Transfers of junior tranches of risk will probably be made harder by regulators, in order to avoid misalignment of incentives. Some kind of resolution of conflicting objectives may also be enforced upon rating agencies.
As latecomers to the process of global diffusion of securitization techniques, Latin American economies may benefit from jumping over pitfalls heretofore known. Given the large scope of unattended needs in terms of finance access and costs in the region, the net balance of securitization is likely to be favorable, as long as lessons from the current crisis abroad are incorporated. Difficult then it will be to contain euphoria!