Session I. The Financial Crisis—Where Are We Now, and What Are the Prospects for Unwinding Public Interventions in the Financial Sector?

Udaibir S Das

Author:

Mr. Udaibir S Das

Mr. Udaibir S Das
Search for other papers by Mr. Udaibir S Das in
Current site
Google Scholar
PubMed

Language:: English

Keywords:: SEM; bank; market; policy coordination; private sector constituent; core element; assessment process; agenda of the Financial Stability Board; Financial sector; Fiscal stimulus; Central bank policy rate; Global; Middle East; Europe; liquidity management policy; emerging market economy; bank liabilities guarantee; bank capital regulation; bank leverage ratio

Download PDF (824.4 KB)

Abstract

The leverage-led growth model—a combination of excessive leverage in the financial system, overindebtedness of households, low interest rates, and global imbalances—was at the heart of the crisis.

Hervé Hannoun Bank for International Settlements

But the paradox is that the policies that have been adopted to remedy the crisis consist, all in all, of even more of the same: borrowing, debt, leverage.

Let me illustrate this with a few facts regarding the main balance sheet adjustments under way:

There has been some reduction of household debt but it still remains at a very high level (Figure 1).
Leverage of banks also remains high by historical standards despite a reduction in the first half of 2009 (Figure 2).
The decline in commercial bank intermediation has been more than offset by the sharp rise in central bank balance sheets (Figure 3). While interbank claims of BIS reporting banks have shrunk by $3 trillion since early 2008, central bank balance sheets have surged from $3.5 trillion to around $7 trillion. This has cushioned the decline in the growth of private bank credit, which turned negative in the last quarter. The near-zero interest rate policy conducted by G-10 central banks has also supported the maintenance of private debt but poses the risk of spurring risk-taking.
And last but not least, there has been a colossal surge in public debt in advanced economies, by 20 percentage points of GDP in two years, to almost 100 percent of GDP in 2009.
Overall, taking decelerating private debt and accelerating public debt together, major economies are still leveraging up (Figure 4). Private debt seems to be still rising in relation to GDP well after government interventions. The ongoing surge in public debt in relation to GDP leaves aggregate leverage (total debt) on a rising trend in many advanced countries.

In a nutshell, we are implementing the leverage-led growth model while promising to break with this model in the future by designing sound medium-term frameworks for fiscal consolidation and bank capital regulation. Promising to “be virtuous, but not now” is a perilous balancing act for policymakers.

Timing and speed of unwinding

The current debate on “exiting too soon” versus “exiting too late” echoes the debate a year ago on the calibration of the stimulus deemed necessary to counter the recession (the risk of “not doing enough” versus the risk of “doing too much” and of overcalibrated stimulus). The difficulty at this point is that, while the recession is abating and the recovery is gaining pace, there is a question mark over the exact measurement of the large-scale stimulus that is in the pipeline and therefore a question mark over the calibration of the stimulus. It may well be that the combination of central banks’ balance sheet expansion and government debt issuance will more than compensate for private credit retrenchment (Figure 4), resulting in overcalibrated stimulus. Another key uncertainty for policymakers arises from the well-known fragility of the measurement of output gaps. As was the case in the 1970s, we may be seriously overestimating economic slack. Mismeasurement of output gaps and growth potential by a wide margin may lead to an understatement of the underlying deterioration in fiscal positions (Figure 5).

The dominant view is that “it is too soon to be implementing the exit strategy but not too soon to be planning for it” or, to be more specific, that “it is too soon to implement any fiscal, prudential, or monetary tightening.” I think that our views should be less categorical.

On fiscal policy

The dominant view (“it is too soon to tighten”) is highly questionable. Postponing the fiscal adjustment to a time when the recovery has consolidated may not be possible.

Inaction and postponement could prove to be a risky policy. Simply communicating on the design of future medium-term frameworks for consolidation may calm the rating agencies for a while, but it does not address the mounting concerns in the bond markets about fiscal solvency in the medium to long term.

On interest rate policy

The dominant view (“it is too soon to tighten”) may be right. But central banks need to keep open the option of starting to reverse the near-zero interest rate policy at any point in time so as to avoid any perception of unconditional commitment to keeping interest rates very low indefinitely. Otherwise market participants will take the current easy financial conditions for granted and start speculating again. Central banks therefore need to make clear that they are adding weight to the “risk-taking channel” of monetary policy and that they will not accept a return to financial excesses.

On central banks’ unconventional balance sheet policies

The dominant view emphasizes the risks associated with the premature withdrawal of unconventional balance sheet policies. Here we need to distinguish between central banks’ short-term liquidity-providing measures and their large-scale outright purchases of long-term securities.

The short-term liquidity-providing facilities can be self-unwinding (a number of them having a fixed expiration date in 2010) and do not pose major exit problems. The pace of that unwinding should be linked to confirmation of the normalization of the LIBOR-OIS spread and a smooth return to private credit intermediation.

Exiting from the outright asset purchases will be more challenging. Given the potential impact on asset prices, central banks may be tempted to adopt a “buy and hold” stance. However, the key issue here relates to the potential role of central banks in directly influencing long-term bond yields and credit spreads: market participants should be under no illusion that we are entering into a new permanent accommodative monetary policy regime in which central banks would be able and willing to control the entire length of yield curves as well as credit spreads and mortgage rates. The unconventional measures should not be seen as an additional set of tools that central banks would use in their normal day-to-day conduct of policy.

In normal times, central banks will need to go back to their usual approach of controlling only the short end of the yield curve and of refraining from interventions with potentially distorting effects on relative asset prices. Exiting from unconventional monetary policy is necessary to make clear that the unconventional will not become the new normal. The sooner the exit, the better.

On prudential policy

On prudential policy, the consensus view again is that “it is too soon to tighten capital requirements”: stronger capital requirements for banks are to be phased in as financial conditions improve and the economic recovery is assured, with the aim of implementation by end-2012. This medium-term phasing-in adopted by the Basel Committee and the G-20 addresses the concern over whether banks would be able to continue their financial intermediation function of providing stable flows of lending. In the meantime, there has been a market-driven increase in banks’ Tier 1 capital ratios of around 2 percentage points between end-2006 and end-June 2009 (Figure 7). But we should not draw too much comfort from this improvement since we know that credit losses in banking lag the business cycle.

There is at least one area where the postponement of a tightening of capital requirements is simply not defensible: the trading book. The additional capital charge there needs to be implemented by the end of 2010 given the extremely low current level of capital requirements on the trading book, even relative to banks’ economic capital estimates.

To conclude on the timing and speed of unwinding: as a minimum, we should recognize that a premature exit and a late exit can be equally damaging.¹

Beyond that, experience suggests that the biggest risk is exiting too late and too slowly or, in the case of fiscal policy, not exiting at all. The political economy pressures are overwhelmingly in that direction. There are three serious risks associated with the policy of “doing nothing now”:

On the fiscal side, the “do nothing” stance could fuel the concern over fiscal solvency, with the potential to trigger bond market disruption. In some G-7 countries, sovereign CDS spreads are as high as bank CDS spreads (Figure 8). It would be more prudent to start the fiscal consolidation effort in 2010 already.
On the interest rate policy side, in addition to the medium-term inflation risks posed by excessive stimulus, the biggest risk in the short term is related to asset price misalignments. The combination of near zero policy rates in G10 countries and excessive risk-taking could create a series of asset price bubbles. Indeed, many observers are concerned by the ongoing resumption of carry trades on currency markets induced by the large interest rate differentials between advanced and emerging market economies (Figure 9). This calls for monetary policy to take better account of asset prices and credit booms, as the BIS has long been advocating.²
On the prudential policy side, the biggest risk to long-term financial stability and sustainable economic growth would arise if the regulatory reform of banks’ capital and leverage were sidetracked.

Some representatives of the banking industry have raised objections to the planned strengthening of capital requirements and the introduction of a leverage ratio as a supplement to the risk-based capital requirement framework to contain the build-up of leverage in the system.

I know that the regulatory authorities will remain firm on these two fronts. Some banks don’t seem to “get it,” and are still promising returns on equity of 20 percent or above to their shareholders. Excessive leverage and risk-taking can no longer be the way to deliver on such promises. They must be grounded in sound, sustainable business models that are robust over a full cycle.

To summarize, we need to embed exit strategies in an overall financial stability framework ensuring consistency across all the elements: unwinding of exceptional fiscal and monetary policies and strengthening of macro- and microprudential approaches to financial supervision.

International coordination of unwinding

Let me finally move to the discussion of the elements of unwinding in which international coordination is essential.

The exit from monetary and fiscal stimulus will have to take into account domestic economic conditions and will therefore be essentially a national decision. That said, G-20 governments have been signaling coordinated fiscal stimulus for the past 18 months. This suggests that coordinated signals should symmetrically be expected in the direction of fiscal consolidation. On the monetary policy side, where decisions are also national, the interest rate cuts announced jointly by central banks on October 8, 2008 were an unprecedented collective action with a powerful signaling effect.

The intensity of the cooperation among central banks was also reflected in the reciprocal bilateral swap lines established to address cross-border foreign currency liquidity shortages. These arrangements will expire early next year, and a smooth coordinated unwinding can be expected.

More generally, the central banking community will continue to make use of existing cooperation forums—especially the committees hosted by the BIS (e.g., the Committee for the Global Financial System and the Markets Committee)—to share information and perspectives on the unwinding of unconventional interventions.

Regarding exit from government financial sector support, international coordination is crucial because of the global nature of large and complex financial institutions and the knock-on effects on other countries.

This is particularly the case for deposit insurance. Disengaging from extraordinary depositor protection measures requires strong international coordination, a good example of which is the initiative taken jointly by Hong Kong SAR, Malaysia, and Singapore to coordinate the exit from the full deposit guarantee.

Another area requiring strong international coordination is the removal of guarantees on wholesale bank liabilities, which may not be a smooth and easy process. The introduction of these measures in the urgency of the crisis was poorly coordinated internationally as regards the fee structure, pricing, and subsidy element of the guarantees.

Among financial institutions, there are differing degrees of dependence on public support. Weaker banks continue to rely on government facilities, while stronger institutions are again able to fund themselves in the senior unsecured bond market. Although it is justifiable for markets to discriminate among financial institutions and to rank (tier) financial institutions according to their dependence on public support, those processes carry a risk: they might in turn lead to pressures toward a new wave of consolidation and even more concentration in the financial sector, thus aggravating the “too big to fail” problem. This means that prudential supervisors have to do everything they can to deal with weak institutions by continuing the process of disposing of bad assets, raising capital, and downsizing where necessary.

In addition, international coordination among market regulators could be useful in directing special attention in this period to the integrity of the information related to financial institutions, ensuring that the tiering assessments floated in the markets are based on solid facts and disclosures and not on rumors or stigma. International coordination in the domain of market integrity and fair competition within the financial sector is also essential to counter any temptation toward “financial protectionism” or promotion of national champions. Finally, internationally agreed prudential standards and their coordinated implementation are essential to ensuring a level playing field, as is the role of the Financial Stability Board in promoting coordination among the standard-setting bodies involved in global regulatory reform.

Malcolm D. Knight Deutsche Bank

The global financial crisis of 2007-09 and the “Great Recession” have propelled three key facets of economic and financial policy—monetary, fiscal, and regulatory—into uncharted territory simultaneously:

The new architecture of financial regulation is still an unfinished project. Recent proposals have not been closely coordinated internationally, creating large areas of uncertainty that make it difficult for financial institutions to design appropriate medium-term business strategies.

Intended and unintended consequences

As expected, policy interest rates at zero and the huge stock of central bank liquidity have led to a rather steep yield curve. However, unintended consequences of the policy stance are already evident:

Zero policy interest rates in key economies, sticky exchange rates against the U.S. dollar, and the rapid growth of central bank liquidity have led to a highly correlated rise in the prices of global equities, junk bonds and—in regions less touched by the crisis—real estate.
As risk appetite has returned, risk spreads, volatility, and other market-based measures of risk have declined.

Sequencing of the exit from the current extraordinary stance of economic policies

The first important point to keep in mind is that the ability of financial institutions and markets to adjust to the authorities’ exit from their extraordinary initiatives depends fundamentally on the creation of an appropriate environment of macroeconomic conditions—one that stabilizes expectations of sustainable noninflationary growth.

The second point to remember is that fiscal sustainability is the essential element of sound economic governance over the longer term.

Therefore, in the long term, a sustainable fiscal stance—one that does not require continuous increases in debt or in the tax burden as a proportion of GDP—is the bedrock of sound macroeconomic management. And the expectation that governments will implement sound macroeconomic policies is the ultimate anchor that stabilizes expectations and thus financial markets.

Given those points, here is the sequencing of exit steps as I see it:

Step one

Create confidence that the fiscal positions of the United States, the United Kingdom, and other countries that are currently running unsustainable deficits will be brought back to a sustainable level in the not-too-distant future.

To anticipate, the second step—normalization of monetary policy—is not feasible without the first step. Unless there is a credible fiscal plan, raising policy interest rates will cause market-determined longer-term rates to rise, increasing the debt-to-GDP ratio. This would be a very negative signal for financial markets.

But even if a forceful adjustment plan to achieve fiscal sustainability can be announced soon and is seen by the markets as credible, the challenges for monetary policy will still be enormous.

As a backdrop to step two, the authorities will need to maintain policy interest rates as low as possible for as long as possible without causing inflation expectations to become unanchored.

Step two

Make a major shift to exit from current monetary policies of quantitative easing, credit easing, and purchasing of private sector financial assets to liquefy key (segmented) markets.

Step three

Establish a credible macroprudential regulatory agency or “financial system risk regulator” in the United States and other key jurisdictions. The agency must have two key features:

the responsibility and accountability for identifying changes in the level of systemwide financial risk; and
the authority to require financial institutions to create countercyclical buffer mechanisms, which mitigate systemwide risk by building higher capital buffers and stronger liquidity cushions in the upswing of the credit cycle and allowing them to run down in periods of financial stress.

Step four

Establish an internationally harmonized intervention and resolution regime for systemically important cross-border financial institutions. The regime must shift the burden of bearing financial losses to unsecured creditors and shareholders to motivate them to monitor financial risks actively. The blanket guarantees on bank liabilities that were introduced during the panic in the fourth quarter of 2008 can then be eliminated.

Step five

Transform the current ad hoc system of swap facilities among key central banks into a permanent, collateralized, multicurrency Lombard facility.

The institution could be based on a permanent system of swap facilities among the central banks responsible for the key currencies. It would have consistent rules on collateral and haircuts to allow, for example, the European Central Bank (ECB) to lend dollars overnight to banks in the European Union or the Federal Reserve to lend euros to U.S. banks in need of overnight foreign currency liquidity.

Step six

In other countries, take appropriately coordinated policy actions to support this sequenced exit by the countries at the heart of the recent financial crisis with the following similarity and difference:

Fiscal policy actions to address rising deficits and debt should have the same priority as those described above for the United States and the United Kingdom.
The monetary and exchange rate policies should be different. Central banks in China and other emerging market economies should implement policies that allow their real exchange rates to rise. For most of these countries, currency appreciation would be a more efficient means of achieving the objective than allowing inflation rates to accelerate.

China should also use this opportunity to undertake a sequenced liberalization of capital transactions and to strengthen the capital adequacy of its banks, both of which would tend to mitigate the upward pressure on its exchange rate.

Step seven

Phase in internationally harmonized reform of the global architecture of financial regulation in a way that avoids a credit crunch but limits the scope for regulatory arbitrage across jurisdictions.

Antonio de Lecea European Commission

Public support has been essential in supporting the economies stricken by the crisis. Addressing a successful exit strategy will be extremely challenging, as the economies are now interconnected, and the risk of policy spillover is significant. The European Union (EU) has taken the lead in defining and communicating a framework for unwinding public interventions in the financial sector. It combines a credible withdrawal of the support measures and good international coordination.

The need for a credible exit strategy

The design of a credible exit strategy and its clear signaling to the market is crucial before actual withdrawal of the support measures. This strategy is essential to anchor expectations and build the public confidence necessary for effective financial, fiscal, and monetary policy interventions. Even if the banking sector has been stabilized, it is not yet able to stand alone without government support. Therefore, to avoid distortion or inappropriate risk behavior, markets should receive a clear signal that public support will not be maintained beyond what is absolutely necessary. Exit strategies need to address several, possibly conflicting, objectives such as sustaining the recovery, rebuilding a stable and viable financial sector able to sustain lending without state support, and strengthening potential growth while tackling macroeconomic imbalances.

The EU exit strategy

The EU is leading the way in designing and developing public awareness of a credible fiscal and financial exit strategy and has a well-established policy coordination framework based on both rules and peer pressure. EU countries will coordinate their various exit strategies through the existing regional budgetary framework, that is, the Stability and Growth Pact. Moreover, the EU Economic and Finance Ministers agreed on the main principles which should underlie the exit strategies: (1) reinforcing incentives to return to a competitive market, (2) ex ante exchange of information on the intention to phase out, (3) transparency toward the public and the financial sector, (4) set the timing on the basis of an assessment of the stability of the financial system (various macroeconomic and financing conditions), and (5) applying the initial phasing out to the general guarantee schemes.

The EU’s coordination framework, though by no means perfect, has shed light on both the issues and the processes relevant for coordination. Even though the initial positions of the various countries were often different, and even though the recovery path and pace may vary across countries and regions, the case for coordination of exit strategies is robust and has been underlined by G-20 leaders.

Specificities of this crisis

Compared to the effort required in past crises, the design and implementation of strategies for unwinding public interventions in the present case has been and will remain particularly challenging, for three main reasons. First, the crisis has been truly global—the degree of interconnectedness and the risks of spillovers across countries are substantially higher than in previous crises as a result of increasing integration. Second, in several countries, today’s rise in deficits and debt comes on top of comparatively high starting points for the ratio of government debt to GDP; in many cases, the withdrawal of the fiscal stimulus and cyclical recovery will not be sufficient to prevent government debt ratios rising to even higher levels. Third, the problem of sustainability generated by the sizeable fiscal deficits is compounded in many countries by the pension and health care effects of aging, which soon will start to hit hard.

As a consequence of these three issues, it is difficult to estimate whether the eventual impact of the crisis will be similar to those of previous crises. There is, for instance, a very high degree of uncertainty about the impact of this crisis on potential growth, about the valuation of assets and liabilities, and hence about the expected losses in the expanded public sector balance sheet.

Timing and coordination are essential

Given the interconnectedness, both in terms of geographic and policy areas, the timing of the phasing-out strategy will be crucial. It should be contingent on the situation in the economy, the financial markets, the health of the individual financial institutions, and possible national specificities. The identification of appropriate timing should also take into account the exit from other support measures, such as the fiscal stimulus and the conventional and unconventional support provided by the central banks.

In particular, several elements must be monitored: (1) macroeconomic conditions and stability (as reflected by indicators such as growth, insolvencies, unemployment, monetary conditions, external developments); (2) banks’ behavior as it relates to a return to “normal market conditions,” that is, adequately easy access to private financing; (3) the strength of the banks’ balance sheets and their capacity to deal with impaired assets; and (4) the actual performance by the financial sector of its role in allocating capital in the economy and ensuring a proper functioning of the credit channels.

Georges Pineau European Central Bank

I would like to thank the organizers of this conference for having invited me to take part in this very timely event. The views expressed here are mine and should not be construed as fully reflecting the positions of the institution I represent.

I will try to address the specific questions put forward by the organizers.

What are the key interdependencies among various types of intervention measures that might make unwinding a complex matter?

Unwinding of policy measures taken in the midst of the financial and economic crisis will be complex because of both their exceptional scope and scale, most notably in advanced economies.

As regards scope, the full gamut of policy areas has been mobilized to support impaired financial sectors and sharply contracting economies. Financial sector measures included mainly bank liabilities guarantees, recapitalization, and asset relief schemes. On the fiscal front, policy steps intended to support aggregate demand, while mitigating financial systemic risks, have resulted in a significant transfer of risks from private to sovereign balance sheets. On the monetary side, central banks’ liquidity management policies used to supplement dysfunctional interbank markets or specific segments of capital markets, have also been reflected in a sizeable “leveraging” of central bank balance sheets. One key interdependency linking these remedial actions stems from the fact that the more these exceptional fiscal and monetary measures are applied on the demand side, the more they undermine incentives to restructure financial institutions on the supply side.

As regards scale, these policies have been stretched to the limits. Specifically, the operation of rule-based fiscal and monetary policy frameworks has been temporarily, but significantly, altered through the use of unconventional measures. This exceptional degree of discretion will have to be phased out gradually as financial and economic conditions stabilize. Such phasing out should be used by relevant authorities to signal that policy modes are gradually shifting as conditions are “normalizing.” Decisions taken by the ECB Governing Council on 3 December 2009 to initiate a gradual phasing out of its enhanced credit support measures illustrate how adjustments in policy modes may be used to “validate” improving financial conditions. In addition, timely exit is needed both to anchor market expectations and to restore policy room that may be required to face further unexpected shocks. The effectiveness of exceptional measures ultimately depends on the credibility of steady-state rule-based policy frameworks. This implies that relevant authorities must map out early enough a reversion to “normal” modus operandi. As regards the risk of additional shocks in the period of rehabilitation, the experiences of countries having faced serious financial dislocation confirm that it is crucial to keep enough policy space, at any point in time, to be able to sustain lasting recoveries.

Which intervention measures have the greatest potential for cross-sectoral and cross-border spillovers?

With respect to the financial sector, the potential for spillovers and distortions exists in two areas in particular.

As regards the first area, various financial sector support measures are likely to impact capital flows or distort competition. For example, blanket guarantees, such as deposit insurance schemes or guarantees of banks’ other liabilities, are likely to affect capital flows, depending on the degree of asset substitution or cross-border financial integration. More targeted measures, such as recapitalization or asset relief schemes, are more likely to distort competition. With a view to mitigating these risks, coordination has taken place, to varying degrees, at the level of the European Union (EU) as well as at the G-7 and G-20 levels. Until now, these initiatives have proved effective in preventing any material rise in financial protectionism.

With respect to the second area, regulatory and supervisory reforms under way in major advanced economies need to be mutually consistent given the cross-border dimension of the ongoing financial crisis. However, such an outcome may prove difficult to achieve for at least three reasons. First, while further regulatory and supervisory convergence is being fostered by the relevant multilateral forums (e.g., the Financial Stability Board, the Basel Committee on Banking Supervision), residual divergences might well persist for some time in certain areas (e.g., capital requirements, accounting standards).

Second, beyond divergences in rule setting, another potential source of regulatory and supervisory arbitrage arises from the lasting tension between supranational rules or standards, on the one hand, and national enforcement and accountability frameworks on the other hand. Third, the supervisory overhaul encompassing micro- and macroprudential arrangements both in the United States and the EU is bound to improve relevant authorities’ ability to identify and address incipient systemic risks. However, it remains to be seen whether these enhanced policy frameworks will contribute to greater global financial stability given prevailing differences in policy constraints and preferences between the two economic areas.

Which crisis policies, when unwound in an uncoordinated manner, bear the greatest downside risk of distorting capital flows and financial intermediation, or of regulatory arbitrage?

Distortions are likely to arise not only as a result of uncoordinated exits but also if exceptional measures are not removed in a timely manner. The EU may be used as a test case given its high degree of financial integration.

Steps have already been taken to restructure banks that had benefited from public support. The first objective of such reorganization is to prevent competitive distortions within the EU single financial market. A second objective consists in reducing the fiscal cost of policy intervention. While the EU rules governing state aid measures ensure consistency within the single market, it is not clear whether an adequate degree of convergence is currently secured among the EU and other large economic areas (e.g., bank resolution activities in the United States).

Further steps that have just been initiated relate to the gradual removal of exceptional financial support. As recommended by the EU Council, blanket guarantees on bank borrowing should be the first type of exceptional support to be gradually phased out, as access to funding markets continues to improve. The first objective of such sequencing is to restore normal market functioning for most financial institutions, while addressing persisting problems in individual financial institutions with appropriate prudential resolution tools (i.e., recapitalization, asset relief schemes). The second objective of this approach is to avoid overburdening fiscal and monetary policies as conditions gradually normalize and macroeconomic instruments lose their usefulness and effectiveness. In addition, by reverting to rule-based macroeconomic policies when appropriate, the EU would facilitate timely exits by other G-20 members.

Conclusions

Conditions for exiting exceptional policy support are gradually falling into place. As systemwide support measures (e.g., liquidity management measures, blanket guarantee schemes) lose their usefulness and effectiveness for a large majority of financial institutions, their orderly phasing-out will be warranted. Timely withdrawal is essential to underpin medium-term credibility of rule-based macroeconomic policy frameworks. At the same time, support measures targeted at individual institutions should remain in place for the time being, as they are required to preserve financial stability.

Manuel Sánchez González Bank of Mexico

During the past two years, governments and central banks have undertaken unprecedented measures to face the risks stemming from the global financial crisis. Most measures were interpreted as extraordinary actions to avoid an economic and financial collapse. At the same time, it was accepted that some costs and risks would need to be controlled. These include tax burdens and possible unintended consequences such as distortions of market participants’ behavior because of reinforced moral hazard. Due to these costs and perils, exit strategies have become a crucial issue.

This conference is exceptionally timely as it addresses the issue of unwinding public interventions in the financial system. In my comments, I will focus on some questions related to the adoption of adequate exit strategies from the perspective of an emerging market economy (EME), with reference to the Mexican experience in the recent and previous crises.

Before moving to the core of the subject, an initial warning seems in order. Exit strategies should not be used as a means to return to a “business as usual” scenario. Just as important as disengagement by the public sector from extraordinary interventions is the government’s engagement in order to prevent future crises. Essential tasks to achieve this goal include sound macroeconomic policies and a revised regulatory and supervisory framework for financial institutions.

EMEs and the global crisis

The recent global financial crisis did not initially have a major impact on EMEs. At a certain stage, a commonly held view was that these economies were decoupling from developed countries. Financial systems in EMEs were solvent and profitable, with high domestic net interest margins having dissuaded local banks from investing in more risky assets. At that stage, the impact of the crisis in EMEs worked through the real side of the economy: a decrease in international trade, a fall in commodity prices, and lower remittances.

However, the worsening of the global turmoil after the Lehman Brothers bankruptcy had major effects. Investment inflows suddenly shrank, and the massive asset sell-off that followed had an adverse impact on exchange rates, domestic interest rates, and local stock markets. Subsidiaries of foreign banks in EMEs constituted an important source of liquidity for their parent firms, which in some cases restricted credit in host countries. Some evidence suggests that since the end of 2008, credit growth in EMEs with a large foreign bank presence has been smaller than in other EMEs.

The financial instability and contraction of output and employment led to the adoption of emergency measures in several EMEs. However, a return to full health in these regions still depends on the recovery of the developed world. The actions adopted in advanced countries to support their financial systems, together with their fiscal and monetary stimuli, have set the stage for the upturn. Proper exit strategies are crucial in order to maintain the improvement of economic prospects.

Preconditions and the unwinding strategy

Governments in many countries have taken numerous measures to support financial systems, including deposit insurance, guarantees of nondeposit liabilities, asset purchases or asset guarantees, special lending facilities and extraordinary central bank liquidity facilities, and capital injections and emergency loans. Exit from these interventions should aim at restoring sustainable financial stability and economic growth.

To achieve a durable and safe disengagement by the public sector, certain preconditions seem desirable. In particular, it is necessary to verify that intermediaries no longer need support, as reflected in their decreased demand for assistance, available access to market sources of funding, and reconstructed balance sheets. This implies the strengthening of capital, reserve, and provisioning ratios under “fair value” conditions. Also, macroeconomic stability should be guaranteed in terms of sound monetary and fiscal policies, so that no undue future financial instability is built into current stimulus measures. On the other hand, economic recovery should clearly be underway, allowing banks to return to normal lending.

Given these preconditions, several elements may contribute to a successful exit strategy. The first one is to price the measures in such a way that a market-based exit is a natural outcome. The incentives of market participants should lead them to draw less on support measures as markets normalize, for example in the cases of debt guarantees, central bank lending facilities, or haircuts to the assets exchanged in a balance-sheet cleanup plan.

A second element is that the timing and sequencing of exit plans should depend on the progress of the preconditions mentioned above, which implies that the timing of withdrawal is partly endogenous and cannot be completely fixed in advance. Although some measures might have been established for a specific time period, others need to be flexible to accommodate unforeseen developments.

A third element is to reconcile the steps of the exit plan with the implementation of any new regulatory and supervisory measures. The combination of higher requirements and lower support should not destabilize banks.

A fourth element is to explore possible cross-border effects so that adequate internationally coordinated measures can be taken. For example, developing countries have been shown to be highly sensitive to relatively smaller changes in the global economic environment. Exit strategies from firm-specific actions may affect countries where those firms play a significant role.

A fifth element is an adequate communication of plans to the public to avoid unnecessary surprises and gain social support. The advance notice of exit strategies, their objectives and timelines, will facilitate adjustment in the regions that expect some indirect impacts or spill-over effects. Transparency and time consistency greatly increase a plan’s credibility.

These elements uncover interdependencies among measures that may make the unwinding process particularly complex. For example, highly accommodative monetary policy in place for too long may stimulate carry-trade transactions that would cause asset price bubbles in EMEs; removal of public sector support without enhanced regulatory rules may facilitate excessive risk-taking; and lack of international coordination may lead banks and their creditors to arbitrage facilities in different countries.

Mexico’s experience in times of crisis

Let me now turn to Mexico’s experience with public sector interventions during the past two crises, which may be useful to policymakers in other countries. The 1995 crisis brought about a sharp depreciation of the peso, a drastic rise in inflation and interest rates, and a deep recession. As a result, the capacity of borrowers to honor their debts was severely impaired, which caused a deterioration of banks’ balance sheets. The authorities responded with a series of measures to stabilize the financial system.

In the wake of the crisis, a dollar-credit window was established at the central bank to help banks service their obligations and reduce the volatility in a highly illiquid foreign exchange market. This facility charged a high dollar interest rate to ensure that the resources were used only for temporary liquidity shortages. The outstanding amount of dollar loans from the central bank peaked in April 1995, and by September of that year all banks had repaid them in full. Thus, the high rates charged on the facility gave financial institutions an incentive to exit, allowing for the facility’s quick termination.

At the same time, to avoid panic, the government announced a blanket guarantee on banks’ liabilities. With the establishment of the new deposit insurance agency (IPAB) in 1999, the coverage was gradually reduced, and by 2005 it had reached its current coverage level (in U.S. dollars) of about $130,000 per depositor and institution.

Also, a program was implemented for banks to sell their nonperforming loans (NPLs) in exchange for the injection of new capital. The deposit insurance agency bought two pesos of NPLs for each peso of new capital injected by shareholders. The NPLs acquired by the agency would still be managed by the selling bank. The scheme included a loss-sharing agreement under which, after 10 years, 30 percent of the losses would be assumed by the banks and 70 percent by the deposit insurance agency.

Unfortunately, measures to fully capitalize the banking system and build conditions for a return to lending activities took several years. One reason was that NPLs were acquired with promissory notes that were nontradable, and, hence, banks could not sell them to finance lending. More importantly, capitalization was limited by legal restrictions on foreign ownership of banks, which were not completely abolished until 1999. Finally, it took three more years to approve reforms to improve the protection of creditors’ rights.

In contrast, during the recent crisis, the emergency measures were milder and, in general, regarded as temporary, with the imposition of explicit deadlines and limits on the amounts of resources committed. An important difference from the previous crisis was the relative strength of the current banking system.

Specifically, in the foreign exchange market, the Bank of Mexico conducted extraordinary interventions as well as daily auctions of dollars with and without a minimum price. The amount offered in the daily auctions was gradually reduced to zero in the case of those with no minimum price. Additionally, the Bank of Mexico and the Federal Reserve agreed on a currency swap line for up to $30 billion, with a deadline extended to February 2010. Finally, the IMF granted a one year contingent credit line to Mexico for about $47 billion. Both facilities are good examples of internationally coordinated measures.

On another front, the sharp steepening of the yield curve had negative effects on institutional investors’ portfolios. To address those effects, the Bank of Mexico conducted interest rate swap auctions. Also, the government and the IPAB reduced the placement of medium- and long-term securities and increased the placement of short-term issues. At the same time, the central bank introduced an auction mechanism to acquire IPAB bonds and implemented the buyback of long-term debt. These measures have begun to see a gradual unwinding. In addition, regulations were amended so that, for a period of six months, investment funds were allowed to carry out purchases and sales of government debt securities with any financial firm belonging to their financial group.

Finally, government-owned development banks launched temporary guarantee programs to facilitate the refinancing of commercial paper issued by businesses and by nonbank, non-mortgage-related financial firms for up to 50 percent, a program that closed to new users in July 2009. These guarantees were adequately collateralized, and their pricing reflected the firms’ debt ratings. Another guarantee program was implemented for nonbank mortgage financial institutions for up to 65 percent, which will conclude in 2010.

In short, in comparison with the previous crisis episode, emergency measures in the recent crisis were considerably less extensive mainly because Mexico had a more resilient financial system, which, in turn, reflected a stronger and more efficient regulatory and supervisory framework as well as improvements in banks’ risk management and macroeconomic fundamentals.

Concluding remarks

In conclusion, careful consideration of the timing of exit from the various public sector interventions is needed. The magnitude of the global crisis, the interdependency of implemented measures, and the needed reforms of the financial system will likely make the exit a long process. Given that support measures may have undesirable moral hazard effects, it is imperative to implement new rules of the game that induce responsible risk taking in the future. This effort should devote special attention to minimizing the size of the “too big to fail” problem, a subject that deserves a whole conference in and of itself. Clear leadership and social consensus around exit strategies and pending reforms will be essential.

Lorenzo Bini Smaghi, “Monetary Policy in Challenging Times,” speech, London, November 19, 2009.

Abstract