Where to from here?

Seven years after the beginning of the crisis, much has been achieved to improve the regulatory framework of the financial system. After the scare of a possible financial meltdown, authorities around the world wanted a simpler system that would be less leveraged, more resilient, transparent and efficient. Nevertheless, several measures are not finalised, implementation of others has been slow and some others will enter into force only in a few years. Referring to the title of this panel “Where to from here?”, I would say that we are not yet “here”. In my remarks, I will start by highlighting some of the achievements and then move on to comment on what, in my view, is still missing.

In broad terms, the first area where major improvements were achieved concerns capital and liquidity regulation: the increase in capital requirements, the introduction of a (still-to-be- calibrated) leverage ratio and the introduction of the two liquidity ratios. These apply to the regulated part of the system, mostly banks. In the US, the banking sector increased by the consolidation or transformation of major broker-dealers into banks. This was not the case for the European banking sector, which actually shrank in the aftermath of the crisis.

It became apparent that what had emerged in the run-up to the crisis was a new market-based credit system funded by short-term secured transactions, protected by risk transfer operations in the OTC derivatives market (IRS, CDS, FXS). This new system had split traditional credit intermediation into separate transactions and was not captured by the statistical system, be it flow of funds accounts or monetary statistics. It was indeed a “shadow banking” sector that was not fully on the radar of regulators. As a result, “shadow banking” is a second area where important measures have been taken and a number of others are under way for adoption in the course of next year. Securitisation, repos and OTC derivatives operationally underpin the development of this new credit system. Securitisation standards have been improved by the application of new regulations and the market has abandoned many unsound practices. The definition of standards for high-quality securitisations has continued and a revision of capital charges is being prepared to re-launch the market on a healthier basis.

Some measures are still to be completed, e.g. on shadow banking or OTC derivatives. The implementation of several others has either been delayed or will take some time to be finalised.

Regarding the repo and securities lending markets, a major step forward was the Financial Stability Board’s (FSB) recent publication of rules for the introduction of minimum haircuts on transactions basically involving corporate securities that are not centrally cleared. This workstream will be concluded early next year.

Concerning institutions belonging to the “shadow banking” sector, new recommendations were issued by IOSCO and, in 2015, an assessment will be made of the implementation of new regulations in different jurisdictions. Progress has been uneven and in several cases insufficient, in my view. Other types of institution will also be reviewed next year by the FSB.

Still in the field of “shadow banking”, regarding the OTC derivatives markets, efforts have been concentrated on transferring the settlement of transactions to central clearing institutions and on stimulating the use of trade repositories to gather and disseminate post-trade information on the transactions. However, implementation was not made mandatory in many cases and implementation of the new standards is well behind schedule. The announced transfer of standardised transactions to organised multilateral platforms is even more delayed as only three jurisdictions have introduced trading transfer requirements for a limited type of derivatives.

A third domain of regulatory reform concerns the overcoming of the “too-big-to-fail” problem. Significant steps forward have been: the increase in capital reinforced by the G-SIB surcharge; the harmonisation of key attributes for resolution already legislated in many jurisdictions; the formalisation of resolution plans that have to be approved by the authorities; the forthcoming introduction of a total loss-absorbing capacity (TLAC) that includes equity and a layer of “bail-in-able debt”; and finally, the adherence of all the major players in the derivatives market to a new ISDA agreement to forgo the termination of cross-border contracts in the event of a globally systemic bank entering resolution.

This was the maximum that could be achieved to address the risks of ring-fencing. But many other aspects will have to remain open, since the harmonisation of national bankruptcy laws is admittedly unrealistic in the near term. The success of the new approach in moving from a culture of bail-outs to a culture of private bail-in depends also on the credibility of the implementation in the eyes of market participants. Another word of caution against over optimism refers to the possibility that problems may not only relate to one institution being “too-big-to-fail” but also possible instances where there are to “too-many-to-fail”.

Another important change in the framework to deal with systemic risk has been the establishment of macro-prudential authorities endowed with a set of instruments to counter risks of financial imbalances, triggering asset price boom/bust cycles.

The range of views about what has been achieved is broad: some say that we are close to overregulation, while others will maintain that nothing fundamental has changed. They are both wrong. It has to be acknowledged that there are important delays in the implementation of some measures, while several others are not yet finalised. This is particularly true of “shadow banking” entities and activities. The recent evolution of the financial system confirms that after an initial decline after 2008, the expanding role of the market-based credit system called “shadow banking” is progressing. In Europe, the banking sector’s total assets have decreased by 11% since 2012, whereas the total assets of investment funds have increased by 30%. A broad concept of “shadow banking” now represents 63% of bank assets in the euro area and more than 100% in the US.

Entities like special investment vehicles (SIVs) or conduits have receded while new forms of funds have flourished, such as: REITs, ETFs, leveraged loans, some CDO structures, and “covenant-lite” loans.

This brings me to the core part of my intervention: what, in my view, is still missing or “Where to from here?” for financial regulation. I will concentrate on five key points: “shadow banking”; OTC derivatives; liquidity mismatches; the macro-prudential toolkit; and incentives in financial institutions.

I. Shadow banking

The first point to consider concerns the insufficient information about the new market-based credit system I just described. These are non-bank institutions that, in a broad sense, perform credit intermediation. This is not bad per se, but the transfer of activity to the “shadow banking” sector entails new risks that are not being monitored. Therefore, it is unclear whether the impact on the overall risk of the system is positive or negative. Even if overall leverage can become less of an issue, liquidity risks stemming from the maturity transformation may considerably increase. Take the case of investment funds, the overwhelming majority of which is of the open-ended type. As investors incur possible losses, there are as such no issues with excessive leverage. There is, however, a significant question of liquidity risk as investment funds’ units are redeemable in the short term and their assets have much longer maturity. The same applies to money-market funds or ETFs. Redemption risk has to be better addressed.

“Shadow banking” is more than just non-bank entities. Shadow banking activities encompass activities often carried out by banks. These include repos and securities financing transactions and the use of different types of derivative. Part of these activities is not covered by existing statistics, justifying the designation of “shadow banking”.

Flow-of-funds accounts record exposures but not the risk transfer generated by derivatives and become therefore misleading. At the same time, forms of quasi-money, such as secured short-term instruments that now represent and act as money in the “shadow banking” sector, are not included in monetary statistics.

We need to create a new statistical apparatus and extend our knowledge of this new sector.

A second point: I would highlight the management of the boundary problems between regulated and less regulated sectors. Even before the crisis, the general mantra was that non-banks did not require any regulation because it was enough, from the regulated side, to supervise banks’ exposures to non-banks. This was not effectively done before the crisis. And at present, the two Basel decisions, namely on the equity investments of banks in non-bank financial institutions and on the limits to large credit exposures of banks, still do not fully acknowledge the existence of “shadow banking”. Large exposure limits relate to exposures to individual clients and not to any particular sector, notably the shadow banking sector. The Dodd-Frank Act however imposes specific quantitative limitations.

Furthermore, the FSOC in the US has the competence, which has already been activated, to declare any institution as financially systemic and therefore subject to close surveillance by the Federal Reserve. This is a powerful tool to apply to the regulated side, light- or non-regulated entities, which does not exist in other jurisdictions, notably in Europe. In my view, the question of managing the boundary problems has to be appropriately addressed through financial regulation.

The third point I want to make concerns the regulation of money market funds (MMFs), an important component of the “shadow banking” sector. Some recommendations on MMFs have been issued by international bodies, particularly IOSCO. There is however still no clarity about what to do in terms of international standards. In the US, the approach taken was to move to variable net asset value (NAV) MMFs. In my view this is an inadequate solution. If Gary Gorton is right in his recent book ^[1], variable NAVs can more easily trigger runs. David Scharfstein has also criticised this approach, which I hope will not be followed in Europe.

The fourth aspect relates to the repo market, which is an essential component of the new market-based credit system. A change in the bankruptcy laws exempting repos from automatic stays in default cases (safe harbour clause) triggered the large increase in the repo market’s size. The regulation of minimum haircuts on such transactions when not centrally cleared addresses an important issue. The issue of re-hypothecation and re-use of securities in securities financing and repos that create chains of inside (or endogenous) liquidity and facilitate rises in leverage is also being examined by a new FSB workstream. Such inside liquidity creates an illusion of liquidity that disappears in stressed times. The excessive reliance on repos and the subsequent “run on repo” were important aspects of the financial crisis. Limitations to re-hypothecation were introduced in the US to respond to this problem. This has however not been done in other jurisdictions, notably Europe. It is expected that the FSB work under way will come up with appropriate recommendations.

II. OTC derivatives

OTC derivatives play an important role in the new credit system by helping to create “relatively safe” assets by transferring or hedging risk. The move to central clearing and to trade repositories, which has been mostly voluntary, has improved transparency and safety in a significant way.

The implementation of the intended transfer of transactions to organised markets of sufficiently standardised derivatives however, has been modest. Very little has been done outside the US, where only a segment of the interest swap market has been affected. Obvious candidates to such transfers to organised markets would be certain types of standardised CDS. The present market is dominated by half a dozen institutions and what is known about transaction prices is based on quotes provided by the major players. In 2009, Myron Scholes said in a conference: “We should blow up or burn CDSs […] and start all over again”. The move to multilateral organised platforms would provide price transparency, reduce transaction costs and increase competition. It should therefore be kept on the regulatory agenda.

III. Liquidity mismatches

An important regulatory change introduced after the crisis was the liquidity regulation with the two new liquidity ratios: the LCR and the NSFR. Both ratios have been watered down since the original proposals, and the LCR, in particular, might have become too lax.

Another relevant aspect is that both ratios are static in nature and not sufficiently granular to facilitate liquidity stress testing or to constitute an overall measure of an institution’s liquidity position. Brunnermeier et al. ^[2] have proposed a new liquidity mismatch indicator (LMI) that would attribute liquidity weights to the relevant items of assets and liabilities. It would provide a sector-wide LMI and enable liquidity stress testing. It goes without saying that more information would have to be collected from the banks and non-banks, but that goes along with the necessary enhanced monitoring of liquidity in the system.

IV. Macro-prudential policies

The regulatory reform brought to the forefront macro-prudential policy as a set of regulatory instruments focused on the overall situation of the financial system. Most of the instruments are of micro-prudential nature, but applied at system-wide level, which lends them a different dimension. Macro-prudential tools are now part of the ECB’s policy toolkit since it was entrusted with supervisory tasks.

The objectives of the new policy include both the improvement of the financial system’s resilience and the smoothing of the financial cycle, as measured by credit and asset prices developments. Empirically, it has been demonstrated that the financial cycle is normally longer than the business cycle, which implies that they evolve differently during certain periods. Testimony to this is the “Great Moderation” period before the crisis, or the present situation. Monetary policy is directed to deal primarily with the business cycle and thus to variables like inflation and output. The general environment of low inflation and low growth which, particularly in Europe, has been recently associated with the hypothesis of secular stagnation, requires an accommodative monetary policy. At the same time, the resulting regime of low interest rates may contribute to fostering a search for yield and creating froth in some asset prices.

This requires central banks which are entrusted with safeguarding the stability of the financial system to be provided with an effective set of macro-prudential policy tools. The ECB has at its disposal the instruments foreseen in the Capital Requirements Directive and Regulation (CRD IV/CRR). These are concentrated on the banking sector. However, any froth that may exist in some European asset markets is, this time, not fuelled by bank credit, which continues to decline. The reduction in bank credit is being offset by the expansion of capital market-based financing or the enhanced role of non-bank institutions. The ECB does not have macro-prudential instruments to deal with the new market-based credit system. The incoming review of CRD IV/CRR may offer an opportunity to reinforce the toolkit by including other instruments such as loan-to-value (LTV) or debt-to-income (DTI) limits, for instance, addressing risks stemming from real estate, or a more extensive use of large exposure regimes (targeting exposures to sectors rather than individual borrowers). The reality is that without more instruments, including those that the Federal Reserve already has, the ECB can hardly be made fully accountable for financial stability in the euro area.

V. Remuneration incentives in financial institutions

The recent distressing revelation of malfeasance in several financial institutions is the most deplorable consequence of the wrong set of incentives that has prevailed in the system. Already in 2005, Raghuram Rajan in his Jackson Hole paper ^[3] pointed to the destabilising role played by the bad incentives that fostered excessive search for yield and leverage. The aforementioned regulatory changes introduced since 2008 are clearly not enough. I agree with the New York Fed President’s recent proposals to introduce deferred compensation (up to ten years) in the form of performance bonds subject to bail-in measures in case of the institution having to pay fines for misbehaviour or having to go into resolution.

Addressing an audience of top managers, Dudley ended with a harsh warning in case incentives did not change in financial institutions: “If that were to occur, the inevitable conclusion will be reached that your firms are too big and complex to manage effectively. In that case, financial stability concerns would dictate that your firms need to be dramatically downsized and simplified so that they can be managed effectively”.

This statement illustrates the multiple dimensions that restoring confidence in finance implies. It is not just all about increasing the capital of the institutions. Many other aspects must still be taken care of if we want to ensure a transparent, trustworthy and efficient financial system.