Credit Institutions
and Procyclicality


Borka Babic, Financial Markets

INTRODUCTION

Lenders have a tendency to amplify cyclical fluctuations, i.e. to have a procyclical effect. This has also been the case in recent years. Due to the crisis and the economic downturn, credit institutions have suffered large losses, which has weakened their capital structures. Many have reacted by tightening their credit standards, e.g. by reducing lending.

A number of factors, including financial regulation, may affect the lend ing capacity of credit institutions over a business cycle. Accounting stand ards and capital adequacy rules are of paramount importance. Among other things, accounting standards determine when losses on loans can be booked. The capital adequacy rules lay down capital requirements and influence the buffers set aside by the credit institutions in good times, which may be used against losses in bad times. The size of these reserves affects not only the robustness of the sector, but also the ability of credit institutions to provide credit in a recession.

Several of the initiatives launched as a result of the financial crisis are aimed at reducing procyclicality. Internationally there is agreement that it is necessary to adjust both accounting standards and capital adequacy rules with a view to reducing procyclicality in the financial sector. Other initiatives include new remuneration rules in the sector and restraints on expanding the credit institutions' balance sheets in relation to their equity (leveraging).

This article describes the relationship between procyclicality and re gulation in the various areas and provides an updated status of the work currently underway in the fields of accounting and capital ad equacy. Initiatives relating to remuneration and leverage in the finan cial sector will also be touched upon.

PROCYCLICALITY: CAUSES AND EFFECTS

As a general rule, growth in lending by credit institutions is strong during booms, while credit standards are tightened in recessions. The boom in the years up to the crisis was thus characterised by a surge in loans relative to the gross domestic product, GDP, cf. Chart 1.1

LENDING BY BANKS AND ECONOMIC DEVELOPMENTS IN DENMARK
Chart 1

chart 1

Source: Ministry of Finance, Statistics Denmark and banks' financial statements.

In good times, when unemployment is low and corporate earnings high, it is natural that credit institutions are more willing to lend. Con versely, in recessions, when the value of any collateral pledged declines, the risk associated with lending to households and the corporate sector is, all other things being equal, greater. Consequently, the credit institu tions become less willing to lend. High lending growth contributes to higher GDP growth during booms, while tighter credit standards damp en economic activity in recessions. Credit institutions thus have a ten dency to amplify cyclical fluctuations.

A number of factors may add to the inherent procyclicality of the credit institutions. One of them is their risk appetite. This changes over time, and – like other market participants – credit institutions have a tendency to assume greater risks during a boom than during a reces sion.2 This affects their lending and thus also the economy in general.

Risk management practices in the financial sector have often been mentioned as a contributory factor behind procyclicality.3 Risk manage ment models tend to attach most importance to recent data for devel opments in the economy and the financial markets. Short-term-oriented or overly simple risk management systems may lead to underestimation of risk in good times, which in turn leads to too easy credit conditions during booms. And vice versa in recessions. A prerequisite for stable risk management systems is that they are based on data and experience from at least one full business cycle.

Information asymmetry between borrowers and lenders is another basic condition for financial business, but may also be part of the explanation to inexpedient procyclicality. Information asymmetry arises because the credit institution naturally has a less complete picture of the borrower than the borrower has. Consequently, it cannot fully take into account the difference between a creditworthy and a less creditworthy borrow er. In recessions, when credit institutions become more cautious in their lending behaviour and the value of collateral decreases, borrowers with profitable projects may also find it more difficult to obtain loans. This is a further impediment to economic activity.4

Financial regulation may also provide incentives for credit institutions and financial markets so that economic fluctuations are amplified. In this respect accounting standards, including rules on write-downs on loans and valuation of financial assets, and capital adequacy rules are key in relation to procyclicality. These issues will be elaborated on in the fol lowing two sections.

The credit institutions' tendency to expand their activities considerably during a boom and reduce them correspondingly during a recession has a major impact on financial stability. The substantial increase in activities – including lending – during a boom may result in the credit institutions taking on excessive risk and thus becoming vulnerable to an economic downturn. Therefore, eliminating the conditions that may lead to inex pedient and excessive fluctuations in financial activity is also important from a financial stability perspective.

ACCOUNTING

Two types of accounting standards are particularly relevant in relation to procyclicality: those concerning provisions for/write-downs on loans, and those concerning valuation of financial assets.

Provisions for/write-downs on loans
Under the International Financial Reporting Standards, IFRS, impairment charges on loans must be made on the basis of objective indication of loss. In other words, an event must occur that results in an identifiable loss, for example that the borrower fails to make the agreed payments.

The principle of objective indication of loss was introduced in the EU in 2005. In Denmark, it replaced the prudential principle, whereby credit institutions had to make provisions for expected future losses. In other words, provisions were replaced by write-downs, and combined with cyclical developments, the new rules contributed to reducing this balance-sheet item, cf. Chart 2.

BANKS' LOAN LOSS PROVISIONS/WRITE-DOWNS
ON LOANS AND OUTPUT GAP IN DENMARK
Chart 2

Chart 2

Source: Banks' financial statements and Ministry of Finance.

The principle of objective indication entails small write-downs and high earnings during booms. All other things being equal, this boosts the credit institutions' capacity to lend and thus to amplify the economic upswing. On the other hand, an economic downturn may result in a pro nounced increase in write-downs, which reduces the credit institutions' earnings and contributes to reducing their capitalisation. This has been evident in the current downturn. Losses may compel credit institutions to reduce lending and thus to accelerate the slowdown.

Internationally it is agreed that the rules need to be adjusted so as to better reflect expected losses on loans. The first proposals in this respect were published over the summer.

The IFRS are laid down by the International Accounting Standards Board, IASB, which is an independent organisation. The IASB is looking at the possibilities of adjusting the existing write-down rules, and in this connection it has published a Request for Information on the feasibility of an expected loss model for the impairment of financial assets. The IASB plans to publish proposals for amendment of the rules in October 2009.5 Unlike the existing rules, this model takes expected credit losses into account, from the loan is booked until it expires. Throughout the term of the loan, calculations are based on the expected cash flows from the loan (adjusted for expected credit losses). If the discounted value of these flows is lower than the book value of the loan, it is written down. Both positive and negative developments in the expected credit losses are to be entered to the income statement on an ongoing basis.

The European Commission's proposal on dynamic provisioning
In July 2009, the European Commission submitted its proposal for imple menting dynamic provisioning for public consultation.6 According to the Commission, these provisions are to supplement the IFRS impairment charges on loans. They are expected to be implemented in the proposal to amend the EU Capital Requirements Directive that the Commission is expected to table in October 2009.7

Under the Commission's proposal, dynamic provisions are to serve as a kind of buffer. This buffer is built up in good times with a view to cover ing expected losses – in excess of the IFRS impairment charges – in a recession. The purpose of implementing dynamic provisions is to ensure that credit institutions have the necessary buffers to cover losses over the entire business cycle. The Commission therefore proposes that the supervisory authorities calculate the expected losses on and provisions for various asset classes on the basis of historical data over a business cycle. The resulting parameters must then be used by the credit institu tions to calculate the required buffers. In its proposal, the Commission assumes that these provisions will be accepted under international accounting standards and that they will have an impact on accounting profit like the ordinary impairment charges. The proposed model is based on the "Spanish model" for dynamic provisions, cf. Box 1.

DYNAMIC PROVISIONS
Box 1

Since the onset of the crisis, much attention has focused on the "Spanish model" for dynamic provisions. Under this model, Spanish credit institutions not only make write-downs in accordance with the principle of objective indication of loss on loans (specific provisions), they also make provisions for expected losses on loans. So far, Spain has been the only EU member state to operate with provisions of this type.

The Spanish model is a statistical model. Dynamic provisions are calculated on the basis of average historical loss and provision ratios over a business cycle relative to the credit institution's specific provisions at the time. Loans are grouped according to six risk categories on the basis of asset credit risk, and historical loss ratios and provisions are based on sector averages for each of these categories. The Spanish supervisory authorities calculate average credit losses and provision ratios, which credit institu tions apply to their portfolios of loans.


Buffers are built up on the basis of the following formula:

Δ Dynamic provisionst = ∑αjΔCjt +∑βjCjt – ΔSPt

where:

  • j indicates the risk category
  • αj is historical average estimate of credit loss for risk category j
  • Cj is the aggregate portfolio of assets in risk category j for which no specific provi sions have been made
  • βj are historical average Specific Provisions for risk category j
  • SP are specific provisions.

At times when specific provisions for losses are low, dynamic provisions increase, and vice versa. In this way, the credit institution's total provisions for losses in any given period become less cyclical.

The model does not take institution-specific circumstances into account, including differences in the various credit institutions' management and pricing of risk. Com bined with the model's dependence on historical data, this is seen as one of its weaknesses. The model is based on the assumption that the data required to calculate historical loss and provision ratios is available. In many EU member states, including Denmark, this is not the case.

In its proposal for implementation of dynamic provisions, the European Commission presents two alternative formulae for calculation of dynamic provisions. The first one is a copy of the Spanish model. The other one is an adapted version of the model in which the first term of the above formula is omitted. According to the Commission, the second model can be applied if the IASB replaces the existing provision model with the "expected loss model".

The Commission's proposal has raised considerable debate internation ally and nationally, one of the issues being whether the proposal for dynamic provisions will affect the accounting standards. Furthermore, a number of technical issues relating to the model have yet to be clarified. The Spanish model has been developed to comply with the existing pro vision model, which is based on the principle of objective indication, and it is therefore uncertain how the Commission's proposal ties in with the IASB's "expected loss model". Moreover, the proposal has the same weaknesses as the Spanish model, cf. Box 1.

Under the Commission's proposal, provisions are to be calculated on the basis of experience with developments in realised losses over a full business cycle. However, no two business cycles are identical, and unpre dicted events can and will occur. Resultant unexpected losses to the credit institutions should, so the Commission envisages, be provided for by building up sufficient capital buffers. This applies to larger-than-expected losses on e.g. loans, as well as depreciation in the market value of assets, including assets booked at fair value, cf. below.8

Valuation
The IFRS provide credit institutions with a choice of valuation methods, including fair value. This method is used for valuation of assets, such as equities and bonds, held by the credit institution for trading purposes. The fair value method entails that assets are always valued at market price when a market price is available (mark-to-market), while model calculations are applied in the absence of a market price.

Unlike valuation methods which are based on historical cost, the fair value method can have a procyclical effect. Market prices vary over the business cycle. Rising market prices during a boom have a positive impact on the profits and capitalisation of credit institutions and may thus induce them to expand their activities. Conversely, falling market prices may trigger a reduction in activities. As the current crisis has shown, market values may drop sharply, and a fall tends to be self-reinforcing, cf. Box 2.

DOWNWARD SPIRAL IN ASSET PRICES
Box 2

Falling market prices for assets valued according to the fair value method have a negative impact on the credit institutions' earnings and capitalisation. Consequently, they may have to reduce their balance sheets and sell some of these assets, i.e. deleverage. This can amplify the downward trend in prices, thereby leading to further losses for the credit institutions. This self-fuelling effect of the decline in asset prices is illustrated in Chart 3. Market illiquidity may accelerate the downward price spiral. This is one of the reasons why the fair value method has been criticised for not working when markets are under stress or characterised by low liquidity.

DOWNWARD SPIRAL IN ASSET PRICES
Chart 3

The fair value method also has many advantages. It contributes to more true and fair, and also more transparent, financial statements. At their April summit, the G20 countries agreed to retain the principle of fair value. At the same time, they agreed that it is necessary to reduce the complexity of accounting standards9for financial instruments and to improve the valuation standards based on instrument liquidity.

When the markets are illiquid, valuation may be subject to consider able uncertainty. One way of addressing uncertainty in the valuation of instruments and to help investors assess the risk associated with a given financial instrument is to tighten the disclosure requirements. The IASB therefore suggests a strengthening of the requirements concerning dis closure of the assumptions and valuation models applied, and in certain cases also the requirements for stating how the choice of valuation method impacts on earnings.10

Another way of addressing the uncertainty concerning valuation of fi nancial instruments is to establish valuation reserves.11 The Financial Stability Board, FSB, recommends that standard setters and supervisors should explore whether firms should be required to hold valuation re serves to avoid overstatement of income when financial instruments are illiquid and valuation is based on data that is difficult to verify.12When an increase in the value of an asset is subject to uncertainty, the added value will thus not be booked, but will be set aside as a reserve.

CAPITAL REQUIREMENTS

In the wake of the financial crisis, several measures have been launched to strengthen the capital structure of credit institutions and to reduce procyclicality. Key initiatives include building up larger capital buffers. In good times, credit institutions should build up buffers so as to ensure that they are also sufficiently capitalised to withstand a recession.

The existing capital adequacy rules, Basel II
Credit institutions play a special role in the economy, and unlike other business enterprises they are therefore subject to extensive regulation, including capital requirements as a major element. Capital requirements imposed by the authorities are, inter alia, aimed at preventing financial problems within one credit institution from spreading and thus threat ening financial stability. This is the case if an event in one credit institu tion triggers substantial financial losses and/or lack of confidence in other parts of the financial system. Capital requirements ultimately re duce the risk of ending up in a situation where the government may have to consider injecting capital into credit institutions.

In terms of capital adequacy, Denmark, like other EU member states, complies with the Basel II Accord, which is integrated into the EU capital adequacy rules. In Denmark, implementation of these rules began in 2007, when they replaced Basel I. The new rules are far more institution-specific than the old rules. Under Basel II, credit institutions have a choice of several approaches to calculation of capital requirements, and the requirements to a large extent reflect the risk incurred by the indi vidual institution.

As early as in the Basel II preparation phase it was pointed out that the risk-based approach to capital requirements could be procyclical. In con nection with the implementation of the new rules an analysis was there fore initiated to further identify their procyclical effects, and regular monitoring of this area was introduced.13 The financial crisis has high lighted the issue.

The risk-based approach means that the minimum capital requirement tends to fall during upswings, when risk is perceived to be limited. This leaves scope for credit institutions to expand their lending activities. Conversely, the capital requirement increases, and lending is reduced, in downturns, when risk is perceived to be high. Thus the capital adequacy rules may reinforce the cyclical trend.

Within the framework of the Basel II Accord, the credit institutions may, however, adopt a more conservative approach to risk assessment. For example, they must reserve more capital to cover risks on the basis of internal stress testing. Within the regulatory framework, supervisory authorities may also impose extra capital requirements on a credit insti tution if they find that the capital calculated by the credit institution is insufficient to cover the risks incurred. How these options can be ex ploited better is also being considered in connection with the ongoing review of the capital adequacy rules in relation to procyclicality, cf. Box 3.

CEBS' PROPOSAL FOR COUNTERCYCLICAL CAPITAL BUFFERS
Box 3

In July, the Committee of European Banking Supervisors, CEBS, published its proposal for reducing the procyclical element in the capital requirements – i.e. the tendency for requirements to decrease during booms and increase during recessions. The CEBS proposal solely concerns credit risk on the credit institutions' banking activities, in cluding ordinary lending activities.1

The proposal is aimed at supervisory authorities and is intended to help them to assess the adequacy of the credit institutions' buffers. The CEBS proposes that supervisory authorities adjust the probabilities of default, PDs, estimated by the credit institution, which are key to the development in the credit institution's minimum capital requirement under Basel II. The current PDs are to be adjusted by a factor reflecting the gap between the PDs in a recession and the current PDs. The credit institution's capital need is calculated on the basis of these adjusted probabilities. In their assessment of the robustness of credit institutions, the supervisory authorities must compare this capital need with the capital need calculated on the basis of the current PDs.

The proposal contributes to the build-up of larger capital buffers in good times. The buffers are expected to increase during booms and to decrease during recessions.

  1. CEBS (2009). Moreover, the CEBS paper concerns only the credit institutions applying their own credit-risk models for calculating the minimum capital requirement for credit risks (IRB, internal ratings based approach).

Larger capital buffers and better capital quality
The Basel II Accord is only just being implemented14 and is therefore not believed to be the root cause of the crisis. Nevertheless, the crisis has de monstrated that the credit institutions are insufficiently capitalised, and that the incentives to build up capital buffers in good times are too weak. Consequently, there is international agreement on the fundamen tal principle that credit institutions must build up larger capital buffers in good times as provision for bad times. This is important both in terms of their robustness to shocks and in order to reduce procyclicality. Low capital buffers could mean that losses compel credit institutions to re duce their activities, including lending, in order to observe the statutory capital requirements.

The specific design of such capital buffers remains an outstanding issue. The European Commission and the Basel Committee on Banking Supervision, BCBS, are both expected to publish their thoughts on and proposals for capital adequacy rules and procyclicality by the end of 2009.15

In connection with the financial crisis, not only the capital level of the credit institutions has come into focus. The crisis has clearly dem onstrated that the capital structure is also significant when the sector comes under pressure. The crisis has disclosed a shortage of high-quality capital, such as share capital, in parts of the banking sector.16 In April 2009 the G20 countries therefore agreed on the need to strengthen the credit institutions' capital base, not only by increasing their capital buffers, but also by improving the capital quality. The BCBS is consider ing various options for strengthening the quality of the credit institu tions' capital, but primarily emphasises that common shares and re tained earnings should constitute the predominant part of the credit institutions' Tier 1 capital.

OTHER MEASURES TO REDUCE PROCYCLICALITY

In addition to the above, a number of other measures have been pro posed to reduce procyclicality. It is thus agreed that restrictions should be introduced on the credit institutions' expansion of their balance sheets in relation to their capital (leverage) that characterised both the global and the Danish financial sector in the years leading up to the onset of the crisis, cf. Chart 4. The European Commission is expected soon to present a proposal for implementation of a simple leverage ratio. This measure is expected to supplement the other capital require ments, either as an indicator to be used by supervisory authorities to assess the sustainability of the credit institutions' expansion of activities, or as a binding requirement.

LENDING LEVERAGE IN THE DANISH BANKING SECTOR
Chart 4

Chart 4

Note: Lending leverage is defined as loans relative to equity capital. The Danish banking sector comprises banks in the Danish Financial Supervisory Authority's groups 1-3.
Source: Banks' financial statements.

Other factors that may have a procyclical effect include remuneration policies in the financial sector. In some credit institutions, remuneration schemes for management and employees have had excessive focus on short-term performance. Moreover, remuneration has frequently been independent of risk-taking. Such schemes therefore provided incentives to expand activities, e.g. in the financial markets, where short-term gains could be reaped by assuming large risks.

In its recommendation of April 2009, the European Commission calls for alignment of pay incentives with the long-term profitability of the credit institutions and for risk adjustment of performance criteria.17 Moreover, the Commission has proposed binding obligations for credit institutions to pursue remuneration policies that support efficient risk management. Under the proposal, supervisory authorities should be empowered to impose financial and non-financial sanctions on a credit institution if it fails to observe this principle. Such sanctions include add itional capital requirements.18

CONCLUDING SUMMARY

In the wake of the financial crisis, a number of initiatives have been launched with a view to reducing the procyclicality of the financial sec tor. Larger loan loss provisions and build-up of larger capital buffers in good times are key elements of these measures.

Work within the various areas is at different stages. As regards provi sions, possible solutions are beginning to emerge, although it is still somewhat uncertain how the European Commission's proposed dynamic provisions will tie in with the IASB's "expected loss model". The issue of designing new capital adequacy rules remains open, however, and spe cific proposals are expected to be tabled in the coming months.

Developments have shown that both larger loan loss provisions and larger capital buffers are necessary. At the same time, it is important to maintain transparency in the financial statements of the credit institu tions when implementing the new rules.

The initiatives are not aimed at resolving the existing crisis, but at increasing the robustness of the sector to future shocks and to dampen the tendency for credit institutions to amplify cyclical fluctuations in fu ture. The initiatives undertaken, and the crisis itself, have probably al ready had an effect on the behaviour of credit institutions. Moreover, many credit institutions have received government capital injections. In other words there is time to consider the possible solutions carefully.

LITERATURE

Andersen, Jens Verner, Hanne Lyngesen and Erik Haller Pedersen (1999), Credit Expansion During Two Booms, Danmarks Nationalbank, Monetary Review, 2nd Quarter.

Andritzky, Jochen, John Kiff, Laura Kodres, Pamela Madrid, Andrea Maechler, Aditya Narain, Noel Sacasa and Jodi Scarlata (2009), Policies to Mitigate Procyclicality, IMF, Staff position note, May.

Berger, Allen and Gregory F. Udell (2003), Institutional memory hy pothesis and the procyclicality of bank lending behaviour, Finance and Economics Discussion Series, No. 2, Board of Governors of the Federal Reserve System.

Borio, Claudio, Craig Furifine and Philip Lowe (2001), Procyclicality of the financial system and financial stability: issues and policy options, BIS Paper, No. 1, March.

Committee of European Banking Supervisors (2009), Position paper on countercyclical capital buffers, July.

EFC (2009), Final Report of the EFC Working Group on procyclicality to the EFC/Ecofin, Working Group on Procyclicality, June.

European Commission (2009a), Possible further changes to the Capital Requirements Directive, July.

European Commission (2009b), Proposal for a Directive of the European Parliament and of the Council amending Directives 2006/48/EC and 2006/49/EC as regards capital requirements for the trading book and for re-securitisations, and the supervisory review of remuneration policies, July.

European Commission (2009c), Recommendation on remuneration pol icies in the financial services sector, April.

Financial Stability Board (2009), Report of the Financial Stability Forum on addressing procyclicality in the financial system, April.

FSA (2009), The Turner review, A regulatory response to the global banking crisis, March.

IASB (2009a), Fair value measurement, Exposure Draft ED/2009/5, May.

IASB (2009b), Financial Instruments: Classification and Measurement, Exposure Draft ED/2009/7, July.

Joint FSF-BCBS Working Group on Bank Capital Issues (2009), Reducing procyclicality arising form the bank capital framework.

The High-Level Group on Financial Supervision in the EU (2009), Chaired by Jacques de Larosière, Report, 25 February.


[1] See Andersen et al. (1999).

[2] See Joint FSF-BCBS Working Group on Bank Capital Issues (2009) and Berger and Gregory (2003).

[3] Cf. e.g. Andritzky et al. (2009) and Financial Stability Board (2009).

[4] Borio et al. (2001).

[5] For further information, see IASB, Request for information: ("Expected Loss Model") Impairment of Financial Assets: Expected Cash Flow Approach, June 2009.

[6] Dynamic provisions have also been proposed by the High-Level Group on Financial Supervision in the EU (the de Larosière Group) (2009).

[7] For further information, see the European Commission (2009a).

[8] All balance-sheet items are comprised by the European Commission's proposal for dynamic provisions, except the trading book.

[9] IASB (2009b).

[10] IASB (2009a).

[11] EFC (2009).i

[12] Financial Stability Board (2009).

[13] Cf. e.g. article 156 of Directive 2006/48/EC of the European Parliament and of the Council of 14 June 2006.

[14] The USA is expected to implement the Basel II capital requirements in 2010.

[15] BCBS, Comprehensive response to the global banking crisis, press release, September 2009.

[16] For an illustration of the impact of the banks' capital structure, see Box 17 in Danmarks National bank, Financial stability, 1st half 2009.

[17] European Commission (2009c).

[18] European Commission (2009b).
Go to buttom
Publication in PDF-format
 
PC: Press the right mouse-button, choose "Save Link As", then choose where to save the file.
 
MAC: Hold down the mouse-button, choose "Save Link", then choose where to save the file.
 
Download
Acrobat Reader here:

 
 
 
Go to previous chapter               Go to top              Go to next chapter