Financial risks: a view from the BIS

Speech by Mr Malcolm D Knight, General Manager of the BIS, at the Central Bank of Argentina 2007 Money and Banking Conference, Buenos Aires, 4-5 June 2007.

Abstract:

In this speech, Mr Knight considers global financial conditions, focusing on: low and stable inflation, which has been associated with rapid credit growth and low interest rates; the proliferation of risk transfer instruments; the emergence of "new" players; and increased portfolio diversification. Mr Knight analyses two factors that may account for increased leverage and how it may unravel. He also discusses developments in Latin America, which has enjoyed a prolonged period of growth and improved resilience to shocks. But vulnerabilities remain. In particular, global financial conditions and growth could worsen abruptly. Moreover, public debt in some countries remains high and is biased towards short-term liabilities.

 

Full speech:

I am delighted to have the opportunity to address this distinguished audience, almost 72 years to the day since the Banco Central de la República Argentina (BCRA) began its operations. The archives of the BIS show that the very first Governor of the central bank, Dr Ernesto Bosch, after whom the hall in which we are gathered today is named, wrote on 20 August 1935 to inform the BIS of the creation of the BCRA and to initiate contacts between the two institutions. The relationship between the BIS and the Central Bank of Argentina has been close ever since.

Today, I will review some international financial developments that have become increasingly evident in recent years; discuss the key risks associated with them; consider various views about the current circumstances and how they might evolve; and comment on some broad risk management issues that are of particular relevance to Latin America at the present time.

1. The global financial environment

Much financial market commentary in recent years has focused on four interrelated developments.

First, in an environment of sustained low and stable inflation, rapid growth of money and credit has been associated with low long-term real yields on financial instruments and a pronounced increase in leverage in the financial sector, which in turn has led to ample growth in global liquidity and a generalised compression of risk spreads. While it is difficult to know how much leverage is appropriate in the financial sector at any given point in time, there is no escaping the fact that current levels of leverage are a source of risk.

Second, credit intermediation has been vastly facilitated by the proliferation of complex risk transfer instruments, including credit derivatives and various types of asset-backed securities. One consequence is that a growing number of banks have been shifting to “originate and distribute” business models, transferring risk to other investors. However, banks’ assets have also increased sharply in recent years, suggesting that they have also kept sizeable complex risks on their own books.

Third, in recent years there has been a sharp increase in the number of “new” players in financial markets, such as hedge funds and private equity firms. The balance of risks from this development remains unclear. These new players, and their use of leverage, have certainly contributed to an increase in market liquidity, but it remains to be seen how they will react in less benign conditions.

Fourth, there has been a much increased diversification of portfolios. Many investors now hold assets that they would not have thought of holding only a few years ago. For instance, emerging market securities and commodities have become an important component of global portfolios. While greater diversification lowers portfolio risk, an unavoidable consequence of this diversification is that it has become increasingly difficult to know who the ultimate holders of risk are and how the prices of specific asset classes will evolve going forward. Moreover, while global investors may be attracted ex ante by certain assets for diversification purposes, the prices of these assets may well turn out ex post to be more correlated than expected with global financial prices.

Greater diversification (new products, new players) has augmented the ability of the financial system to assume risk. And not surprisingly, these changes have also led to lower risk premia and an increase in actual risk-taking in a range of markets, a development that has been amplified by the fact that longer-term interest rates began declining in both nominal and real terms in the mid-1990s and have been at historically very low levels in the last few years, particularly relative to actual GDP growth rates.

Housing finance

One market in which these forces have come together to increase risk-taking is housing finance. Across the world, the availability of mortgage credit has improved, household borrowing has grown substantially, and housing prices have risen, increasing the exposures of the financial system to real estate conditions. These developments appear to be largely due to households reacting to declining interest rates by increasing their demand for housing and mortgage credit. Though interest payments have so far remained broadly stable relative to income, increased leverage makes households more sensitive to both interest rate and income fluctuations. In addition to these greater macroeconomic risks, those posed by the sheer complexity of new financial instruments have also risen. To continue with the example of the housing market, increasingly complex residential mortgage contracts have been introduced. These have tended to hold down interest payments during the first few years of a mortgage at the cost of raising them in the future, either because of “teaser” rates or simply because a common feature of these contracts is that the interest rate is variable and tends to be low in situations in which the general level of interest rates in the economy is low.

As a result of innovations like these, households that previously were not seen as creditworthy have been increasingly able to obtain financing. While better access to finance can be a good thing, the pool of borrowers may have become riskier, as the share of lower-income and less financially strong borrowers has risen. It is easy to see that the combination of a potentially riskier pool of borrowers and mortgage contracts entailing more interest rate risk can raise financial stability concerns.

Leveraged financing

Lower interest rates and a decline in interest rate volatility have also been associated with a generalised rise in leverage and more risk-taking in the financial sector. These developments have gone hand in hand with a rapid expansion of markets for complex and risky assets – including leveraged loans, 1 credit derivatives and various structured products – in which hedge funds have become increasingly important suppliers of liquidity. The benign credit environment and very favourable actual and projected corporate earnings have also supported a boom in private equity transactions and a substantial increase in merger and acquisition activity. And while all this has been happening, leveraged loan spreads have actually tightened by more than 100 basis points over the past two years.

However, the embedded leverage in many of these new financial products can be difficult to measure. This also complicates counterparty risk assessment, since it is more difficult to determine who the ultimate holders of risk are. These factors have increased vulnerability to market shocks. In the leveraged loan market, rating agencies have already drawn attention to deteriorating credit quality standards generally and a weakening of loan covenants. An additional source of concern is that the recent very large expansion of financial transactions relies on the ability of markets themselves to function smoothly in times of stress. The provision of liquidity in the face of shocks is therefore an essential element of risk management.

However, let me emphasise that while risks may have risen, risk-taking is a fundamental element of financial intermediation. Moreover, it is not the level of risk per se that is a source of concern, but rather that, in the current benign environment, some categories of risk may not be properly understood, correctly priced or appropriately managed.

2. How financial risks may materialise

To assess the risks to the international financial system, it is essential to distinguish between observable symptoms and their underlying causes. While low real interest rates have provided incentives for economic agents to take on more risk, and have thereby played an important role in the developments described above, the issue of why rates have been so low remains unsettled.

Many have argued that a firmer commitment to price stability by central banks across the world has reduced nominal interest rates and may have contributed to the compression of risk premia. Other monetary policy developments – including improved communication of the authorities’ objectives and greater transparency – are also likely to have played a role in reducing inflation-adjusted yields. According to this view, the principal risk is a large and unexpected tightening of monetary policy, perhaps in response to a sharp rise in inflation.

There are reasons to think that monetary policy has also been an important factor behind the recent benign conditions in financial markets. In particular, the long period of monetary accommodation in many places around the world may have contributed to low real long-term interest rates and upward pressures on the prices of various assets. In addition, concerns have been raised over the consequences of low policy interest rates in many parts of the world and associated “carry trades” involving borrowing in low-yielding currencies to invest in high-yielding currencies but also in countries where asset prices or currency values were expected to increase. However, many central banks, notably the Federal Reserve, have already raised policy rates repeatedly without triggering sharp market corrections. This suggests that monetary policy factors have not been the only ones at play.

Another hypothesis is that a global saving-investment imbalance has developed, as desired savings are high but intended investment in fixed capital is low. As our previous speaker, Raghu Rajan, has observed, this fact may have led to a shortage of new debt securities. 2 As a result, yields have fallen and investors have been pushed into a host of “new” asset classes, raising their prices, depressing risk spreads, and leading to much more diversified portfolios.

And of course another – partly related – factor has been the sharp increase in foreign exchange reserves observed globally in recent years. Massive sterilised intervention has been carried out, contributing further to the maintenance of very loose domestic monetary conditions particularly in Asia. And as the accumulated reserves have been reinvested, they have added to the already ample liquidity in global financial markets.

Although all the factors I have just described appear to have been influential in recent developments, it is difficult to say which factor will be the key determinant in the future, in terms of risk in global financial markets. Since saving-investment imbalances evolve slowly over time, one would expect that the observed changes in yields and spreads will be reversed only gradually. But a further prolonged period of low interest rates and tight spreads of the sort that is implied by the saving-investment hypothesis risks encouraging even more leveraging in the short to medium term. The further build-up of global foreign exchange reserves could have similar implications. This suggests that the current episode of ample liquidity may be longer, involve a continued build-up of positions, and have an even more uncertain resolution than one might at first have expected.

By contrast, if monetary policy has played a dominant role, the rise in inflation that has been observed recently in many countries and the likelihood of a further tightening of global monetary conditions suggest that the current episode of low interest rates and tight spreads could end quickly. This could have an adverse impact on interest rate sensitive sectors of the economy and lead to a withdrawal of liquidity from precisely those markets that have benefited the most from low interest rates.

3. Financial risks in Latin America

Let me now move from a general discussion of global financial risks to the more specific question of the financial risks that could potentially affect the Latin American region. No one can fail to notice that Latin America has experienced remarkable economic expansion since 2004. Thus far the economic upswing has occurred without the emergence of significant disequilibria, which is a major achievement for a region where bouts of growth have traditionally been accompanied by mounting imbalances and, ultimately, economic crises.

Of course, Latin America has recently benefited from favourable external financing conditions, strong global economic growth, and high commodity prices. This environment has helped countries strengthen their fiscal positions, reduce their external debt burdens, shift to more stable debt structures, and accumulate large foreign exchange reserves. All this, in particular sounder macroeconomic policies, has been reflected in sharply lower sovereign spreads. As an illustration of this trend, the spread on the securities included in the EMBI+ index for Latin America has fallen from more than 1,000 basis points at the end of 2002 to about 170 basis points in May 2007. As a result, Latin America now seems better placed to weather financial turbulences than it was a decade ago.

Potential risks related to the global financial environment

However, vulnerabilities remain, reflecting the global factors I mentioned earlier. Ample liquidity and tight spreads have supported credit growth in many countries of the world, and not least here in Latin America. But conditions in financial markets can change abruptly. An unexpectedly sharp tightening of global monetary conditions or a financial shock could lead to a contraction of global liquidity. Indeed, recent episodes of global market volatility remind us that pressures can appear suddenly. For example, a number of analysts have questioned whether the compression of risk spreads across Latin America is fully justified by economic fundamentals. Hence a key issue for the countries in the region concerns the extent to which these economic fundamentals have improved in the recent period of favourable economic conditions. It is hard to answer this question. But clearly the economies in the region that maintain higher credibility for their macroeconomic policy stance are likely to be subject to less stress than others in the face of adverse shocks.

In that context, it should be noted that while much progress has been made in reducing public sector debt burdens in Latin America, several countries in the region still have high public debt – above 50% of GDP in several cases (Table 1). Moreover, the share of short-term or floating rate liabilities continues to be high (Table 2). Clearly, debt sustainability could be affected by a rise in global interest rates and its consequences for growth in Latin America. This is why countries in this region should take advantage of the current favourable global macroeconomic and financial environment to make further progress in ensuring debt sustainability. This, certainly, could be onerous in the short term – moving from dollar-denominated and/or short-term debt to often more expensive local currency denominated and/or long-term debt. Nevertheless, it would probably be a wise decision for the future.

Furthermore, domestic credit is currently growing rapidly in the Latin American region following years of stagnation, reflecting the recovery of local banking systems, but also some rise in domestic investment, a buoyant export performance, and strong capital inflows. Of course, while financial deepening is desirable, the recent surge in domestic credit in a number of Latin American countries has been associated with an increase in inflationary pressures. Given the high level of commodity prices and tightening output gaps, ensuring price stability could become a more challenging task. As history indicates, rapid growth in credit could also create problems at a later stage if lending is not properly managed and supervised.

In recent years, strong terms of trade gains have fuelled growth, helping Latin American countries to replenish public coffers and to rebuild foreign exchange reserves. But the exposure of the region to global growth and developments in commodity prices is still high. Potential output growth remains low by international standards, and progress in developing non-commodity exports has been limited in several places.

This suggests that authorities should continue to take advantage of the currently benign – largely externally driven – financial conditions to implement the adequate policy mix (eg enhanced supervision and regulation, and prudent fiscal discipline) that would ensure better macrofinancial stability.

Local currency bond markets

In this context, let me briefly focus on one important structural development in the Latin American region that seems to us at the BIS to represent a key step towards a better macrofinancial framework. This is the development of local currency bond markets, which have expanded rapidly in the region as a result of policies aimed at reducing dependence on external capital flows and at addressing the vulnerabilities associated with currency mismatches.

Such markets support financial stability by improving the efficiency of financial intermediation, limiting currency and maturity mismatches, and reducing the concentration of risks in the banking system. Certainly, and as always, these new developments also create new risks. Let me mention a few:

These risks are particularly relevant in those domestic bond markets that are still at an early stage of development and lack the features that help more mature markets to cope with periods of stress. From this perspective, a few specific policy goals appear key. Among these would be: first, to ensure that domestic bond markets are liquid enough for exposures to be adjusted rapidly without significant disruptions to market functioning; second, to allow an adequate diversification of risks across sectors, preferably by encouraging private sector issuance and thereby developing needed market infrastructure (eg clearing and settlement, risk markets and risk transfer instruments, and the legal framework of the financial system); and third, to develop better frameworks for the monitoring of risk exposures.

4. Conclusions

Let me conclude. Ample global liquidity, new financial instruments and players, and a marked increase in portfolio diversification have improved the global financial system’s ability to bear risk and market participants’ willingness to assume it. But risk has not disappeared from the system, and it needs to be managed closely and effectively. This applies particularly to Latin America, which is currently enjoying strong macroeconomic conditions and has made substantial and impressive progress towards financial stability, but where financial markets could still face challenging times if global conditions were to deteriorate.

The BIS’s Fifth Annual Report, published in 1935, contained a discussion of the establishment of the Banco Central de la República Argentina. A reading of that report in the light of several further decades of experience suggests that our views of financial risk have not changed so much in the intervening period. Let me conclude with a quote from another chapter of that report (p 40): “[t]here is a common desire … that more care should be taken by borrowers and lenders in scrutinizing the purposes for which financing is sought. But it would be vain to think that a complete elimination of risks can be achieved; in financing as in other human activities, no development can take place without some venture.”

Thank you.

 

Table 1: Public sector debt in Latin America
(% of GDP)

 

Domestic public sector debt

External public sector debt

2000

2006

2000

2006

Argentina

15.2

29.2

29.9

28.1

Brazil

40.3

50.6

14.3

7.1

Chile

28.1

8.4

8.0

7.4

Colombia

26.7

33.7

26.3

18.4

Mexico

12.1

18.4

14.6

6.5

Peru

9.5

9.1

35.9

23.6

Venezuela

8.8

10.8

18.6

14.8

         

Latin America

23.1

30.6

18.7

11.3

Source: JPMorgan Chase (April 2007).

 

Table 2: Domestic bonds in Latin America by type of instrument
(% of outstanding stock)

 

2000

2005

Floating

Straight fixed rate

Inflation- indexed

Currency- linked

Floating

Straight fixed rate

Inflation- indexed

Currency- linked

Argentina

12.0

0.0

0.0

88.0

2.0

1.0

74.0

20.0

Brazil

58.0

15.0

6.0

21.0

60.0

21.0

16.0

3.0

Chile

0.0

0.0

92.0

8.0

0.0

18.0

64.0

18.0

Colombia

0.0

50.0

41.0

7.0

0.0

70.0

29.0

1.0

Mexico

35.0

6.0

16.0

0.0

47.0

28.0

13.0

0.0

Peru

17.0

0.0

54.0

29.0

3.0

35.0

36.0

25.0

Venezuela

100.0

0.0

0.0

0.0

44.0

0.0

0.0

56.0

                 

Latin America

47.0

12.0

13.0

22.0

46.0

23.0

23.0

5.0

Source: CGFS Working Group on Local Currency Bond Markets. Data cover only bonds and notes and exclude money market instruments. Regional totals based on the countries listed in the table. Totals do not add up to 100% due to the exclusion of hybrid instruments.


1 Leveraged loans are defined by the IMF as: “bank loans that are rated below investment grade (BB+ and lower by S&P or Fitch, and Baa1 and lower by Moody’s) to firms with a sizable debt-to-EBITDA ratio, or trade at wide spreads over LIBOR (e.g., more than 150 basis points).”

2 See Raghuram G Rajan, “Is there a global shortage of fixed assets?”, remarks at the G30 meetings in New York, 1 December 2006, available at www.imf.org.